Skip to main content

Full text of "On early English pronunciation, with special reference to Shakspere and Chaucer, containing an investigation of the correspondence of writing with speech in England, from the Anglosaxon period to the present day, preceded by a systematic notation of all spoken sounds by means of the ordinary printing types"

See other formats

















PART 11. 








p. 13, 1. 7 from bottom, omit But for convenience, a very brief key is given on 

p. 16. 
Cancel p. 16, which is replaced by pp. 614-5. 

p. 105, n. 2, 1. Q,for b^lt read (b<?lt). o -7 -^ o 

p. 107, 1. 4 from bottom of text, /or (a) read (Ai). I \i f <J ^ 

p. Ill, 1. 6, /or (oi, ou) read (oi, on). ^ 

p. 118, 1. 6 from bottom, /or tEMS rm^ tEms. ^My 4 ^ IQ/^T 

p. 119, 1. 15, for aryl read ray?. "^ ' f 

p. 141, 1. 8 from bottom, omit as we sounded lyke. 

p. 153, 1. 9 and 3 from bottom of text, omit and which, and that the change. 
p. 254, n. 1, 1. 6, omit (possibly a reference to St. Mary le bon) ; n. 3, add at the 

end of this note : See note on v. 672, Chap. VII. \ 1. 
p. 265, 1. 24—26, omit But susteene . . . 8323. 
p. 309, n. 1, 1. 3, for z read g. 
p. 333, 1. 26—29, read " Tyrwhitt, and the MSS. of the Six-Text Edition of 

Chaucer, read thilke for the'^ Omit another mode . . . wikkedly, 
p, 333, n. 1, 1. 8, /or Hengwit read Hengwrt. 
p. 336, n., supply ^ 

p. 347, art. 17, 1. 10, /or -innge read -innge. 
p. 355, art. 5^, for Ex. to (c), read Ex. to {a). 
p. 371, Ex. col. 1, 1. 28, before wiltow insert [c). 
p. 388, after Manhood insert 14. 
p. 407, table col. 2, 1. 4, for "(ou) 00 oa'' read "(00) 00 oaJ' Note that 

" (ou) ou ow" in col, 3, I. 4 is correct. 


p. 473, n. col. 2, 1. 1,/or p. 446 read p. 447. 

p. 477, n. 2, 1. 3, 07nit more. 

p. 506, n. 2, last word, /or (riu'le) read (ruu-le). Seep. 573, under ITI. 

p. 562, translation, verse 13, 1. 4, /or yon, rmc? yonder. 



On account of the unexpected length of the present inves- 
tigations, the Societies for which they are published have 
found it most convenient to divide them into foiir parts, 
instead of two as previously contemplated. The present 
second part concludes most of the researches themselves. 
The third part, containing Chapters YII. and VIII., is in 
the press, and will be ready by January, 1870. Chapter 
VII. will contain an introduction to the specimen of Chaucer; 
a critical text of the Prologue to the Canterbury Tales, with 
especial reference to final e, metrical peculiarities, and intro- 
duction of French words, together with the conjectured 
pronunciation ; a passage from Gower, printed for the first 
time, according to three MSS. with the conjectured pronun- 
ciation ; and a specimen of Wycliffe. Chapter VIII. will 
contain Salesbury's and Barcley's works ; specimens of 
phonetic writing in the xvi th century, by Hart, Bullokar, 
Grill, and Butler ; a Pronouncing Vocabulary of the period ; 
an account of French and Latin pronunciation in the xvi th 
century ; an examination of Spenser's and Shakspere's 
rhymes, and Shakspere's puns ; and an attempt to restore 
Shakspere's pronunciation. The fourth part, will treat of 
English pronunciation during the xvii th and xviii th cen- 
turies, and of dialectic usages, and will contain full indices 
to every part of the work, but the time of its appearance 
cannot yet be announced. 

A. J. E. 


1 Aug., 1869. 

coxte:ms of paet it. 

CHAPTER V. Ox the Pronunciation of English during the Thir- 
teenth AND Previous Centuries, and of the Teutonic and Scandi- 
navian Sources of the English Language. 

§ 1. Rhymed Poems of the xiii th Century and earlier, pp. 417-485. 

No. 1. The Cuckoo Song (with the Music), circa A.D. 1240, 

pp. 419-428. 
No. 2. The Prisoner's Prayer (with the Music), circa A.D. 

1270, pp. 428-439. 
No. 3. Miscellanies of the xiii th Century, from the Reliquiae 
Antiquoe, Early English Poems and Political Songs, 
with an Examination of the Norman French EI, AI, 
pp. 439-466. 
No. 4. The Story of Genesis and Exodus, circa A.D. 1290, pp. 

No. 5. Havelok the Dane, circa A.D. 1290, pp. 470-479. 
No. 6. King Horn, circa A.D. 1290, pp. 480-483. 
No. 7. Moral Ode, Pater Noster and Orison, xiith Century, 
pp. 484-485. 
§ 2. Unrhymed Poems of the xiii th. Century and Earlier, pp. 486-497. 
No. 1. Orrmin's Orrmulum, end of the xiith Century, pp. 

No. 3.. Layamon's Brut, beginning of xiiith Century, pp. 
§ 3. Prose Writings of the xiii th Century and Earlier, pp. 498-508. 

No. 1. Only English Proclamation of Henry III., 18 Oct. 

1258, pp. 498-505. 
No. 2. The Ancren Riwle, xiii th Century, pp. 506-507. 
No. 3. Old English Homilies, xii th Century, pp. 507-508. 
§ 4. Teutonic and Scandinavian Sources of the English Language, pp. 
No. 1. Anglosaxon, pp. 510-537. 

No. 2. Icelandic, pp. 537-553, and Old Norse, pp. 554-560. 
No. 3. Gothic, pp. 560-564. 

CHAPTER VI. On the Correspondence of Orthography with Pronun- 
ciation FROM the Anglosaxon Times to the Present Day. 

§ 1. The Value of the Letters, pp. 565-588. 

^^ 2. The Expression of the Sounds, pp. 589-606. 

§ 3. Historical Phonetic Spelling, pp. 606-618. 

§ 4. Etymological Spelling, pp. 618-621. 

§ 5. Standard or Typographical Spelling, pp. 621-623. 

§ 6. Standard Pronunciation, pp. 624-632. 

For the intended contents of the whole ivork, see Fart I. 



On the Pronunciation of English during the Thirteenth 
AND Previous Centuries, and of the Teutonic and 
Scandinavian Sources of the English Language. 

§ 1. Rhymed Poems of the Thirteenth Century and Earlier. 

It remains for us to apply the method employed for as- 
certaining the pronunciation of English during the xiv th 
century, to the discovery, if possible, of that of the xiii th 
century, and for this purpose it is necessary to examine the 
rhymed poems of this date in manuscripts which seem to 
belong with certainty to that period. Poems composed in 
the XIII th century, but transcribed in the xiv th, and there- 
fore presenting the peculiar orthography of the latter period, 
are of little use for our purpose. This will account for the 
rejection of many rhymed poems which belong to this period. 
The following cases ha^^e been selected with some care. 

The Cuckoo Song and Prisoner's Prayer, which stand 
first, have their antiquity well established, and possess the 
great advantage of a contemporary musical setting, which is 
of considerable assistance in determining the pronunciation 
or elision of the final e. As the old notation of music re- 
quires especial study to read, faithful translations into the 
modern notation, preserving exactly the number and pitch 
of the notes, have been printed. This is precisely similar to 
reducing the manuscript letters to the form of Roman types, 
extending the contractions and pointing. In the first piece 
the time of each note is accurately determined in the original, 
and is strictly observed in the transcript. In the second, 
which is in plain chanty this is not the case, and hence such 
time has been assigned as was suggested by a careful ex- 
amination of the notes in connection with the words. 

In approaching these earlier poems we stand already upon 
very secure ground. The values of a, ai, an, e, ei, eUy i, ie, 
0, oi, OK, as (aa a, ai, au, ee e, ei ai, eu, ii i, ee, oo o, 
ui, oou ou) have every appearance of being the most ancient 
possible, and the only doubtful points turn on such fine 

V ■< 27 


distinctions as (a r/, e e, i ^), which it would be impossible to 
determine from the rhymes alone with certainty, since the 
necessarily strongly provincial character of all early poems, 
will certainly admit of rhymes apparently lax, which only 
represent peculiar pronunciations. In fact there was no 
longer a common or a recognized superior dialect, for the 
English language had long ceased to be that of the nobility. 
From the Anglo-Saxon Charters of the Conqueror down 
to the memorable proclamation issued by Henry III. (see 
below, p. 498j, and for a century afterwards, the English 
language was ignored by the authorities, and was only used 
by or for ^' lewd men.*'^ But there was a certain amount of 
education among the priests, who were the chief writers, and 
who saved the language from falling into the helplessness of 
peasant dialogue. 

The chief points of difficulty are the use of ou for (uu, u), 
the use of u for (yy, y) and even {i, e), and of eu for (yy). 
The meaning of ea, eo, oa, practically unused in the xiv th 
century, has also to be determined. The result of the pre- 
sent investigation may be conveniently anticipated. It will 
be found that ou was not used at all for (uu, u) till near the 
close of the xiii th century, when the growing use of u for 
(yy) or (i, e), rendered the meaning of u uncertain. But in 
the pure xiii th century writings u only is employed for (uu), 
and becomes a test orthography (p. 408). The combination eu 
or ew, does not seem to have been used except as (eu). The 
combinations ea^ eo, so frequently rhyme with e, and inter- 
change with it orthographically, that their meaning was 
probably intentionally (ea, eo), with the stress on the first 
element, and the second element obscure,^ so that the result, 
scarcely differed from (ee') or even (ee). The combination 
oa was either (aa) or (aa). The consonants seem to have 
been the same as in the xiv th century, although j may pos- 
sibly have retained more of the [gh) than the (j) character. 

1 Man og to luuen "Sat rimes ren, which case, according to some writers, 
"Se Wiffed wel ^e logede men, the first element falls into (j, w), which 
hu man may him wel loken however, others deny. In (iu, ui) the 
"Sog he ne be lered on no boken, stress is properly on the first element, 
Luuen god and feruen him ay. as also in most provincial diphthongs 
Genesis and Exodus, 1-5. beginning with (i), as (stiaan, mien) 
])is hoc is y-mad uor lewede men. = stone, mane. But in Italian chiaro, 
Ai/enbite of Intv^t, sni^rk 1^. 4:12. ghiaccio (A;iaa*ro, .^'iat'tshio) the (i) is 
2 The general rule for the stress upon touched quite lightly, and is almost 
the elements of diphthongs is that it evanescent, so that (kjaa-ro, gjat*- 
falls upon the first, but this rule is tsho) would generally be thought 
occasionally violated. Thus in many enough. A method is therefore re- 
combinations with initial (i, u) the quired for indicating the stress, when 
stress falls on the second element, in difliculty might arise, or when it is 

§ 1, Ko. 1. 



1. The Cuckoo Song (with the Music), cmcX a.d. 1240. 

The Harleian MS. 978, in the Eritish Museum, was a monk's 
alhum or commonplace book. It is a small vellum MS. entirely of 
the xm th century, but evidently written by many hands at dif- 
ferent times. The contents are very miscellaneous. It begins with 
several musical pieces, some with and some without words, Latin, 
Trench, and English; it proceeds to give an account of musical 
notation and tones, then suddenly commences a calendar, of which 
only the first two months are complete, though the others are 
blocked in. Then comes a letter to Alexander the Great on the 
preservation of health, Avicenna on the same, account of the 
seasons, melancholy, etc., all in Latin. On fo. 24, the language 
changes to French, and we have recipfes for oxymel, hypocrase, etc. 
On fo. 32, the hand changes, but the recipes are continued. The 
language reverts to Latin on fo. 32^, and the hand changes again 
on fo. 33h, col. 2, line 2. Without pursuing the catalogue further, 
we may notice a change of hand again on fo. 37 and fo. 38, where 
a beautifully written French Esop commences. We have again a 
different hand on fo. 66b, and so on. In the later part of the 
volume is a Latin poem of (twice) 968 lines on the Battle of Lewes, 
14th May, 1264, (printed by Mr. T. Wright in his Political Songs, 
pp. 72-121), in which the cause of the Barons against Henry III., 
is so warmly taken, ^ that it must have been composed, and pro- 
bably also transcribed, before they were utterly routed and ruined 

abnormal, and for this purpose the 
acute accent may be used, as (Aiaa-ro, 
^iat'tshio), and similarly (ea, eo) in 
some theoretical pronunciations of 
anglo-saxon, and this accent may be 
used in all cases if desired. In Ice- 
landic I have heard the triphthong 
(ioou) with the unusual stress on the 
first, and (ie) when apparently (ie) was 
written, and in such cases the mark is 
indispensable. In Icelandic, I have 
also found it necessary to symbolize a 
very faint pronunciation of a letter, 
rather indicated than pronounced, 
rather felt by the speaker than heard 
by the listener, by prefixing i a cut [ , 
to such a letter, as the symbol of evanes- 
cence, so that we might write (eLa) for 
(ea) that is (ea), or (^LiaaTo, ^[iat*- 
tshLio) if preferred. If it is wished to 
shew that a whole word or phrase is so 
spoken, then it should be enclosed be- 
tween L 1 ; thus, clergymen will fre- 
quently faintly indicate words preced- 
ing an accented syllable, as (['n eti 
'keeiim Lt'1 pahs) =and it came to pass. 
These symbols must be considered as 
appended to the list of palaeotypic signs, 
supra p. 12. 

^ Compare the opening lines — 
Calamus velociter 

scribe sic scribentis, 
Lingua laudabiliter 

te benedicentis, 
Dei patris dextera, 

domine virtutum. 
Qui das tuis prospera 

quando vis ad nutum ; 
In te jam confidere 

discant universi, 
Quos volebant perdere 

qui nunc sunt dispersi. 
Quorum caput capitur, 

membra captivantur ; 
Gens elata labitur, 

fideles Isetantur. 
Jam respirat Anglia, 

sperans libertatem ; 
Cui Dei gratia 

det prosperitatem I 
Comparati canibus 

Angli viluerunt, 
Sed nunc victis hostibus 

caput extulerunt. 
Wright prints each pair of lines in one, 
as in the original MS., but the rhymes 
point out this present division, which 
doubles the number of lines in the 



Chap. v. 

at Evesham, 4th Aug. 1265. This is therefore important in fixing 
the date of the MS., but Sir Frederick Madden assigns to the first 
portion of the MS. a date twenty or thirty years earlier, and believes 
that the writer, that is, transcriber, — by no means, necessarily, 
author — was a monk of the Monastery at Reading, founded by 
Henry I, 1125.^ 

poem. It was be seen from these lines 
what smoothness of versification the 
monks in the xiii th century were ac- 
customed to; with only some slight 
accentual liberties, and what perfect 
rhymes they formed in Latin. We 
shall find the same smoothness in a 
very similar metre in Orrmin, and 
hence must expect that the English 
versification of the present period will 
also run without stumbling, unless the 
writer is very uncultivated. 

1 The following notes are written in 
pencil at the beginning of the volume. 
" The whole is of the thirteenth cen- 
tury, except some writing on ff. 15^- 
17. F.M."— "In all probability the 
earlier portion of this volume was 
written in the Abbey of Eeading, about 
the year 1240. Compare the Obits in 
the Calendar with those in the Calen- 
dar of the Cartulary of Eeading, in 
MS. Cott. Vesp. E.V. F.M. April 
1862." Mr. William Chappell has 
kindly favoured me with the inspection 
of a letter from Sir F. Madden, in 
which he gives the grounds for this 
opinion, and as the date of the MS. is 
of considerable importance to our in- 
vestigation I add an abstract of the 
same, which Sir F. M. has politely re- 
vised. 1. It is certain that the first 
part of the MS. (say the first 30 folios) 
is considerably older than the second, 
which contains the poem on the battle 
of Lewes composed 1264. 2. In this 
first part is a portion of a calendar, 
containing the obits of Abbots Koger 
19 Jan. [1164]; Auscherius 27 Jan. 
[1135] ; Eegiuald 3 Feb. [1158] ; 
Joseph 8 Feb. [circa 1180]; and Sy- 
mon 13 Feb. [1226]. In Browne 
Willis's History of the Mitred Parlia- 
mentary Abbies, etc., 1718, vol 1, p. 
159, ail these Abbots are named, as 
Abbots of Eeading. 3. The complete 
calendar, left unfinished in Harl. 978, 
is found [with the exception of Dec] 
in the Cartulary of Eeading, Cotton 
MS. Vesp. E.V. fo. ll^i to fo. 165. The 
latest obit recorded in the old writing 
of the mouths after Feb., is that of Abbot 

Adam de Latebury, 6 April 1238, all 
later obits are in a clearly marked later 
hand. The part of the Cartulary coeval 
with the Calendar was written about 
1240, for fo. 22A contains a charter 
dated 24 Henry III., 1239-40, and at 
fo. 33(5 is a marginal note written sub- 
sequently to the text, and dated 29 Hen. 
III., 1244-5. In Jan. and Feb. the 
obits are the same as in Harl. 978, 
[with this difi'erence that in the Harl. 
MS. Abbot Eoger's obit is given under 

19 Jan., and in the Cotton MS. under 

20 Jan.] From these facts Sir F. 
M. "considers it proved by internal 
evidence, First, that the Calendar in 
both MS." and consequently the pre- 
ceding parts, "was written in 1240 or 
very little later. Secondly, that the 
Calendars . . . were undoubtedly written 
at Eeading, by a monk of that house. 
Lastly," he adds, "there is a remarkable 
entry in the Calendar of Harl. 978 (but 
omitted in thatof Vesp. E.V.) on St. Wul- 
stan's day, 19th Jan., as follows : — Ora^ 
Wulstane, pro nostro fratre Johanne de 
de Fornsete. I am strongly tempted 
to regard this John de Fornsett, (who, 
from his name must have been a native 
of Norfolk), as the Scribe of the MS., 
for I cannot otherwise account for the 
odd introduction of his name in the 
Calendar." The entry referred to is 
literally as follows, the italics indi- 
cating extended contractions : — " xiiu 
kal(?w6?a« Wlstani epjscopi obiit Eogerws 
abb«s. Ora Wlstane pro nostxo iratre 
Johanne de fornfete." The omission 
of the u after JF, as in Wulstan is not 
uncommon, but it is noteworthy in 
this place, because in the Eng- 
lish Song, which will be presently 
given at length, wde for wude occui's, 
and this a priori connects the two 
writers together, but of course the per- 
son who wrote that entry, which is in 
exactly the same handwriting as the 
rest, could not have been John of 
Fornsett. Hence I should consider 
this entry as making it highly probable 
that this monk was not the scribe, 
and the singular insertion may be due 
to his having been an intimate friend 

§ 1, No. 1. 



This MS. contains on fo. lOh. the music and words of the Cuckoo 
Song, which, Mr. "W. Chappell says, ''is not only one of the first 
English songs with or without music, but the first example of 
counterpart in six parts, as well as of fugue, catch, and canon ; and 
at least a century, if not two hundred years, earlier than any com- 
position of the kind produced out of England." ^ This song which 

of the scribe. The MS. was evidently 
one for private use, and this note of a 
friend's death is anything but surprising. 
" You are probably right as to John de 
Fornsete not being the scribe," re- 
marks Sir F. M., "still the introduc- 
tion of his name is very singular, and 
I do not recollect any other instance 
of a, friend heing thus commemorated." 
The above historical external evidence 
of the real date of this MS., is rendered 
the more important because Hawkins 
2, 93, and Burney 2, 405 in their His- 
tories of Music, attribute it to the 
XV th century, "misled," says Sir F. 
M., " by an ignorant note of Dr. Gif- 
ford on the fly-leaf of the volume," and 
by the nature of the musical composi- 
tion, which they supposed could not 
have been written before the time of 
John of Dunstable in the xv th century, 
an opinion refuted by Mr. W. Chap- 
pell, who quotes "Walter Odlington, 
1228-1240 (Scriptorum de Musica 
Medii -^vi novam seriem a Gerbertina 
alteram collegit nuncque primum edidit 
JE. de Coussemaker, Paris, 1863, 4to., 
p. 245) to this effect : " Habet quidem 
Discantus species plures. Et si quod 
unus cantat omnes per ordinem reci- 
tent, vocatur Kondellus, id est, rotabilis 
vel circumductus." We also know 
that the English spelling of Cuckoo in 
the XV th century was Cuckow^ not 
Cuccu, which could only have been 
used in the xiii th. 

1 W. Chappell, F.S.A. Popular 
Music of the olden time, a collection 
of Ancient Songs, Ballads, and Dance 
Tunes, illustrative of the National 
Music of England, etc. The whole of 
the airs harmonized by G. A. Macfar- 
ren. (Printed 1855-9) p. 23. Mr. 
Chappell has given a facsimile of this 
song as the title page to his work, and 
says, in the explanation of that plate : 
" The composition is in what was called 
* perfect time,' and therefore every long 
note must be treated as dotted, unless 
it is immediately followed by a short 
note (here of diamond shape) to fill 
the time of the dot. The music is 

on six lines, and if the lowest line 
were taken away, the remaining would 
be the five now employed in part 
music, where the C clef is used on 
the third line for a counter-tenor 
voice. . . . The Pound has been re- 
cently sung in public, and gave so much 
satisfaction, even to modern hearers, 
that a repetition was demanded." 
He adds in another place, p. 23 : — 
" The chief merit of this song is the 
airy and pastoral correspondence be- 
tween the words and music, and I 
believe its superiority to be owing to 
its having been a national song and 
tune, selected according to the custom 
of the time as a basis for harmony, 
and that it is not entirely a scholastic 
composition. The fact of its having a 
natural drone bass would tend rather 
to confirm this view than otherwise. 
The bagpipe, the true parent of the 
organ, was then in use as a rustic in- 
strument throughout Eiu-ope. The 
rote, too, which was in somewhat better 
estimation, had a drone, like the modern 
hurdy-gurdy, from the turning of its 
wheel. When the canon is sung the 
key-note may be sustained throughout, 
and it will be in accordance with the 
rules of modern harmony. But the 
foot or burden, as it stands in the 
ancient copy, will produce a very in- 
different effect on a modern ear, — 
we ought perhaps to except the lover 
of Scotch reels — from its constantly 
making fifths and octaves with the 
voices, although such progressions were 
not forbidden by the laws of music in 
that age. No subject would be more 
natural for a pastoral song than the 
approach of summer, and, curiously 
enough, the late Mr. Bunting noted 
down an Irish song fi-om tradition, 
the title of which he translated * Sum- 
mer is coming,' and the tune begins 
in the same way. That is the air to 
which Moore adapted the words, * Rich 
and rare were the gems she wore.' " 
This resemblance is perfectly fortuitous, 
and does not extend beyond the first 
three notes, the fourth note of the Irish 



Chap. V. 

is so great a musical curiosity, is also a valuable contribution to our 
knowledge of early English pronunciation. In order to make the 
song more readily legible, it will be here interpreted into the 
ordinary musical notation,^ the English words in Roman type, and 
below them the Latin hymn, by which it perhaps obtained its in- 
troduction into the monk's commonplace book,'^ in Italics, (which 
when used for entire passages will indicate red ink,) and a literal 
translation of the notes into modem music. On the opposite page 
will be given the metrical arrangement, conjectured pronunciation, 
and literal translations.^ See pp. 426, 427. 

air runs into a totally different chord. 
The fact that the song was in six parts, 
has occasioned some persons to sup- 
pose that it was alluded to in the last 
stanza of the * Turnament of totenham,' 
Harl. MS. 5396, fo. 310, the hand- 
writing of which is referred to a.d. 
1456. As the stanza is not printed 
quite correctly in Percy's Reliqiies, 2nd 
ed., ii, 15, it may be added here as 
transcribed from the original MS. It 
is scarcely right to suppose, however, 
that the Cuckoo Song was the only 
six part song known. 
At ])at feft )»ay wer^ fipruyd wetA a 

ryche a ray 
'Etuery .v. and v had a cokenay 
And fo j^ay fat in jolyte al ])Q lang day 
And at \e lafl ))ay went to bed viiih 

ful gret deray 
mekyl myrth t^as ]?em amang 

In Query corner oi\Q hous 

Was melody delycyous 

For to her^ pr(gcyus 
of vj menys fang. 
Dr. Rimbault has published a modern 
version of this song in his Ancient 
Vocal Music of England, Novello, No. 
13, in which he says: "the editor 
has followed an ancient transcript in 
the Pepysian Library, which omits the 
two bass parts forming the burden, 
in the Museum copy, and has added an 
Accompaniment upon a drone bass. 
The effect produced is considerably im- 
proved." Dr. Eimbault has politely 
informed me in a private letter to Mr. 
G. A. Macfarren, that he obtained his 
copy of this transcript from the late 
Prof. Walmisley of Cambridge, in 1838. 
Mr. Aldis Wright kindly made a search 
for the original in the Pepysian Library, 
but was unable to find a trace of it. 

1 Hawkins and Burney (supra, p. 420, 
note 1, near the end,) have given 
translations with all the parts written 
at length, but have not arranged the 

words properly. In the present inter- 
pretation the arrangement of the ori- 
ginal is followed, and for one deviation 
from the former translations I am in- 
debted to Mr. William Chappell. 

2 Mr. G. A. Macfarren, the com- 
poser, in reply to my question whether 
he considered the English or Latin 
words to have been the original, says : 
*'I am strongly of opinion that the 
music was composed to the English 
words, and the Latin Hymn afterwards 
adapted to it, because it was a common 
practice to adapt sacred words to secu- 
lar tunes (as for instance, Thomas, 
archbishop of York in the xith cen- 
tury and Richard Vichys of Ossory in 
the XIV th wrote many such), but it 
would have been regarded as a dese- 
cration to appropriate a church theme 
to a secular subject. Witness also the 
many masses set to music, throughout 
which the French song of VJiomme 
Arme is employed as a canto fermo, 
and Josquin de Pre's Mass on this Song 
in praise of Chess, in proof of this same 
church practice." To this we may add 
that there are no Latin words to the 
Fes or Burden^ which is an essential 
part of the harmony. 

3 This arrangement is reprinted from 
the work cited below, p. 498. As re- 
spects the language, all the words are 
ags. except cuccu, stert, uert. 
The first cuccu as we shall see is 
onomatopoetic (imsonic, or mimetic), 
the second stert, and its diminutive 
startle^ is fully at home in the German, 
old sturzan, new sturzen, and Scandi- 
navian, Danish styrte, Swedish storta, 
and may be a development of stir^ or 
may be related to the same root as ags. 
steortan to erect, steort a tail, steart a 
spine, see Dief Goth. W. 2, 304, 315, 
333, Wedgewood, Etym. Diet. 3, 314. 
As to the third uert, Dr. Stratmann 
suggests fert^ which would be the 


The musical notes, with their precise value in time, and the Latin 
hymn, determine the number of syllables. As we find however the 
Latin accent occasionally violated {non parchis, v'lte dondt et seciim 
corondt), we cannot be surprised at a similar \dolation of the Eng- 
lish, in Wei singes \u. Taking the notes as intei'preted on p. 426, 
it would seem easy to rearrange the words so as to avoid this false 
accentuation, but the ligatures of the original, corresponding to the 
slurs in the translation, forbid this rearrangement, which, with 
other liberties, Hawkins and Bumey have not hesitated to adopt. 
Hence we find that this termination -es, might be, and probably 
was, fully pronounced. On the other hand, the termination -e\y 
although fully pronounced in growe^, blowe]>, was elided, either 
after a vowel or consonant, when convenient for the metre as in 
spring]) ; or for the music, as in Ihou]}. In the latter case the metre 
would require the syllable -e]> to be fully pronounced, compare 

Awe bletej? after lorriJ) 
Loue]) after calue cu, 

but the musician ventured not only to dock a syllable, but to put the 
whole heavy truncated word lhou]> to a short note. This may teach 
us that our older and ruder poets did not hesitate to lay words on a 
Procrustean bed. In medy hulluc, ags. medu, lulluca, the poet took 
the same liberty, and elided the final -e^ for the rhyme in the first 
case, for the metre in the second. This precisely agrees with what 
we determined to be the occasional practice of the xrvth century 
(p. 342, l!^o. 5), and shews that the omission was absolute, not a mere 
slurring over or lightly touching of the sound. "We must consider 
that the words were felt to be as really truncated as Ruh^ for Ruhe 
appears to be in modern German speech, for we have the essential 
-e preserved in ivde, aive, hucke, the dative -e in calue, the adverbial 
-e in Ihude, murie, all of which have a distinct musical note assigned. 
In the last word, however, both vowels in -ie are given to one note, 
as mang a time would be given to three notes only in modem ballads. 
The principal fact, however, that we learn from this song, as to 
the pronunciation of the letters in the xin th century, is that long 
(uu) which was represented generally by ou and occasionally by o, 
but never by u, in the xiv th century, was now invariably repre- 
sented by u. This is deduced from the word cuccu, which is mani- 
festly an imitation of the cry of the bird,^ as in French coucou, old 
French coucoul, Italian cuculo, German liukuk, huclcuh, Dutch hoehoeh 
(kuu'kuuk), Latin cuculuSy coccyx^ Greek kokkv^, Sanscrit hohila.^ 

ags. feortan, pedere, but this change such an interval in connection with the 

of / into «;, although frequent in old cry, being in v. 6, where in sing cuccu 

MSS, is not confirmed by any other he first descends and then ascends a 

usage in the present poem, and the use of minor third, the notes being f df. 

a Norman word vert in a hunting phrase ^ « Cuckoo in Enghsh is clearly a mere 

seems natural. The use of the word as imitation of the cry of that bird, even 

a verb, however, requires confirmation. more so then the corresponding terms 

^ The musical interval of the cry is in Greek, Sanskrit, and Latiu. In 

a descending minor third, which the these languages the imitative element 

composer has not imitated, the only has received the support of a derivative 

instance in which he has introduced suffix ; we have hoM^ in Sanskrit, and 


The sound must have been (kuk'kuu*) or (kwk'kuu*) or simply 
(kz^k'u), as at present. The orthography may be compared with the 
ciichow of Chaucer 17174 (supra p. 305), where the short (u) remains 
the same, but the long (uu) is represented by ow. Agreeing with 
this we have Ihude, mi, cu, \ii which were lowde loude, now, cow, 
thou in Chaucer. And thus the characteristic difference between 
the orthographies of the xmth and xivth centuries (p. 408,) is 
established by reference to a bird's cry, which cannot have changed. 

But u in the xin th century did not always represent the sounds 
(uu, u), as we see by the word murie, which however is not enough 
in itseK, or even when compared with the ags. mirige, to establish 
the second sound of u as (i) or (e), or originally (y) as previously 
suggested (p. 299). In Hali Meidenhad^ we constantly find u for i 
01' y. Thus in the first page, hlv^eluher ags. bli^eHce, blithely, 
lustni ags. lystnan, listen, hrudlac, ags. brydlac, marriage gift, clup- 
pinge ags. clyppan, clip embrace, hwuch ags. hwilc, which, \unche^ 
ags. ]7mt?a?^, seem ; euch each, in which last word the sound (eutsh) 
is almost unthinkable. The town of Hertford is so spelled in the 
Prench version of the English proclamation of Henry III, but 
appears as Hurtford, in the contemporary English version, 1258. 
The conclusion seems to be rather that the u, which was properly 
and generally employed as (uu, u), was coming into use to replace 
the ags. y (y), which it succeeded in doing by the end of the xin th 
century, thereby necessitating the recurrence to ou for (uu). Was 
this double use of u, then, due to the IN'orman influence ? In the 
French version of the Proclamation already cited, ^ we have Cunte, 
tu%, nus, piir, sicum, iurz, sunt, etc., in which u was most probably 
(uu, u), while in Due, saluz, greignure, esluz, iurgent, desuz, etc., 
the sound could hardly have been other than (yy, y). The 
I^orman u derived from Latin u may have been frequently (yy), and 
that derived from Latin o, may have been generally (uu). The 
point is not yet satisfactorily established,^ and the English and ITor- 

hohkyx in Greek, euculus in Latin, as other names for the cuckoo, old Sla- 

(Pott, Etymologisclie Forschungen, i. vonic gz'egz'olka, Lithuanian ge'guz'e, 

84 ; Zeitschrift, hi. 43). Cuckoo is, in Lettish dfeggufe and Lithuanian ku- 

fact, a modern word, which has taken koti, to scream like a cuckoo, old Norse 

the place of the Anglosaxon geac [cfceJc], gaukr (goeoefkr) etc., and gives other 

the German Gauch (gauk^t'h), and, examples of names of birds from their 

being purely onomatopoetic, it is of cry. Cumberland (gauk), Scotch (gauk). 

course not liable to the changes of ^ Hali Meidenhad, from MS. Cott. 

Grimm's Law. As the word cucTcoo Titus D. xviii.fol. 112c ; an alliterative 

predicates nothing but the sound of a homily of the thirteenth century, edited 

particular bird, it could never be applied by Oswald Cockayne, M.A., once of St. 

for expressing any general quality in John's College, Cambridge; published 

which other animals might share ; and for the Early English Text Society, 

the only derivatives to which it might 1866. 8vo. pp. 50. 

give rise are words expressive of a me- "^ Both versions are given below, 

taphorical likness with the bird. The pp. 500-505, accurately printed from 

same applies to cock, the Sanskrit kuk- the originals in the Public Record 

kui:a.'" Max Miiller, Lectures on the Office. 

Science of Language, 1861, p. 347. ^ Mr. Payne is of opinion that the 

Pott, in the passage referred to, gives Norman u, ui, were always (uu). Com- 


§1, No. 1. CUCKOO SO^'G — XIII TH CENTURY. 425 

man orthographies derive so differently, that in the xm th century 
they can scarcely be held to influence each other. Hence the in- 
troduction of ou for (uu) into English may be a native development, 
as already stated, and not due to French customs. The frequent 
appearance of w, where i would be expected, in Western English, 
as in dude, lute for dide, lite, may at most indicate a wider geo- 
graphical extension of that sound (y) which is now nearly con- 
fined in the west to Devonshire. In our inability however to 
determine the last, especially in Eastern and Southern English, 
where we find the orthographies w, i, e interchanging, we have 
no choice but to pronounce as i, e {i, c). See the remarks on the 
same use of u in the xrvth century, supra pp. 298-300. !N^umer- 
ous examples will occur in the following pages of this section. 

We gather then from the Cuckoo Song: 1) that ou, ow were 
used for (oou) only, as in lhou\, growe\, ags. hlowan, growan, and 
never for (uu, u) which were uniformly represented by u, but u 
itself was probably ambigTious, and also represented an actual or 
older (yy, y), which was interchangeable with «', e ; 2) that e final 
was regularly pronounced, but might be suppressed even not before 
a vowel, when required for the metre or rhyme ; 3) that -ep might 
be pronounced or suppressed ; 4) that -es might be so distinctly 
pronounced as to be sung to an accented note. 

As regards the remaining letters and combinations no information 
is given, but on the other hand there is no reason to suppose them 
different from the sounds already obtained for the xiv th century. 
The words are practically the same. The consonants no doubt had 
not altered. The vowels a, e, o had abeady received their most 
ancient powers (a, e, o). The only doubt affects i, which in the 
xrvth centuiy we concluded to be {ii, i). There can be little doubt 
that the Latin value of these letters was (ii, i), but it does not 
follow that when the Saxons changed their runic for the Roman 
alphabet, they actually said (ii, i). If they had said (n, i) it would 
have been near enough. In subsequent examples we shall frequently 
find i, e short confused, which would still lead us to suppose that i 
short was (^ ) rather than (i). But from this time forth the evidence 
is not. strong enough for long i being {ii). It certainly could not 
have been (ai), if we were right in concluding that it was {ii) in 
the XIV th century (p. 297). In this doubtful state of the case, I 
shall adopt (ii, i) as the long and short sound of i, in all my indi- 
cations of the pronunciation of the xm th century and earlier, and 
content myself with recording here once for all that I consider the 
short i to have been certainly {i), and that the time when long i 
passed from {ii) into (ii), if there ever was such a time in England, 
is unknown. Upon these grounds I have drawn up the pronun- 
ciation exhibited on (p. 427). 

pare : bure mesaventure, bure couver- u had almost certainly the sound of (yy), 

ture from King Horn, infra p. 480, and and it is possible that this later ortho- 

the spelling huis muis, p. 449. When graphy may be a guide to the oldest 

the spelling ou was established for (uu), pronunciation. 






From the Harleian MS. 978, /o. 10 b. 


v-mer if i - cu-men in. Lhud-e ling cue -cu. Grow-e|j fed andblow-e|j 
Fer-fpi - ce christ -i - co - la. que dig-na • ci - o. ce - li - cus a - gn - co- 

"h n 

r 1 

1 ] 




1 1 — 1 

V h 



o J 

<^ ■ 

Uvr:, J 

<^ 1 




O 1 

VW «> (d 



med andfpring}? J7e w - de nu. Sing cue -cu Aw-e ble-te{> af- ter 

la pro ui-tif vx - ci - o. ft - li - o non par - cenf ex - po - fu- 


■^ 7^- 




-Q — fS- 



lomb. Ihou)? af - ter cal - ue cu. 
xt. mor - tis ex - z - ex - o 

Bull-ue ftert - e)?. buck - e uert - e\> 
Qui cap - ti - iios fe - mi - ui - vos 












Mu - rie ftng cue - cu 
a fup - plx-cx - — 

Cuc-cu cue - cu "Wei fin-gef Jju cuc-cu ne fwik 
Vi - te do - nat et fe - cum cor -o - nat in ce- 




Hanc rotam cantare poffunt quatuor focij. A paucio- 

;: ribuf aiitem q?fain a tribus uel faltem duobus now debet 



I,,, -r^oiioy T,„ dici. preter eof qui dicuwt pedem. Canittt?* autem fic. Tacen- 

pu nauer nu. * x x- 

Ix fo'li-o tibws cetenfunMsinchoat cum hijfqwttene»t pedem. Etcumuenerit 

ad pWmam notam post crucem s inchoat aliuf. & fic de cetens 
" "^ (T). - ^-. lingwli uero repaufent ad paufacionef fenptas & 

now alibi ' fpacio uniuf longe note— 



Pef ing cuc-cu nu. Sing cuc-cu. 








hoc repetit unns quocienf opns est • 
facienf pavfacxonem (nfine. 

hoc dtcxt alius, paufans fn medio & won (ti 

ing cuc-cu. Sing cuc-cu nu ytne. Sed immediate repetens prmctpixim. 

§ 1, No. 1. 




From the Harleian MS. 978, fo. 105. 

Marly English Original. 

Svraer if icumen in. 

Lhude fing cuccu. 
GroweJ? fed 
and blowef med 

and fpringj? \q wde nu. 
Sing cuccu 
Awe blete]? after lomb. 

lliou]? after calue cu. 
Bulluc ftertef. 
bucke uertej? 

Mune fmg cuccu. 
Cuccu cuccu 
"Wei fingef fu cuccu 
ne fwik fu nauer nu. 

Smg cuccu nu. Sing cuccu. 

Sing cuccu. Sing cuccu nu. 



Conjectured Pronunciation. 

Suu'mer is ikuu'men in. 

Lbuu'de siq, kuk'kuu* ! 
Groou'eth seed, 
And bloou-eth meed. 

And spriqth dhe uud*e nuu. 
Siq, kuk'kuu* ! 
Au-e bleet'eth after lomb, 

Lhoouth af-ter kal've kuu. 
Eul'uuk stert'eth, 
Euk-e yert'etb, 

Mer-ie siq, kuk'kuu ! 
Kuk'kuu* ! kuk'kuu* ! 
Wei siq*es dhuu, kuk'kuu* ! 
Nee swiik dhuu naver nuu. 


Siq, kuk'kuu*, nuu ! Siq, kuk'- 
kuu* ! 

Siq, kuk'kuu* ! Siq, kuk'kuu*, 

Verbal Translation of the Early English. — Summer has come in, Loudly sing, 
cuckoo ! Grows seed, And blossoms mead, And springs the wood now. Sing, 
cuckoo ! Ewe bleats after lamb, Lows after (its) calf (the) cow. Eullock leaps. 
Buck verts (seeks the green), Merrily sing, cuckoo ! Cuckoo, cuckoo ! "Well 
singest thou, cuckoo, Cease thou not never now. Burden. Sing, cuckoo, now ! 
sing, cuckoo ! Sing, cuckoo ! sing, cuckoo, now ! 

Latin Hymn to the same notes. — Perfpice Xp'icola. — que dignacio. — celicus — 
agricola — pro uitif vicio. — filio — non parcenf exposuit — mortis exicio — Qui 
captiuos— femiuiuos — a supplicio — vite donat — et secum coronat — in cell folio. 

Verbal Translation of the Latin Hymn. — Behold, Christ-Worshipper {Christi- 
cola) What condescension ! From heaven The husbandman For the fault of the 
vine, His son Not sparing has exposed To the destruction of death. Who the 
captives Half-alive From punishment Gives to life, And crowns with him In 
heaven's throne. 

428 prisoner's prayer — xiii th century. Chap. v. 

Three peculiarities -will here be noticed (au'e, lomb, naver), 
corresponding to aive, lomh, naver, in the MSS. Since, then, the 
sciibe is supposed by Sir F. Madden to have been a Norfolk man, 
I endeavoured to write the song in the present !N'orfolk pronun- 
ciation, and having submitted the following to competent revision 
I believe that it is sufficiently correct to shew that if the old pro- 
nunciation, already given (p. 427), has any claim to consideration, 
there is no ground to suppose that the song was written in an 
East Anglian dialect. The East Midland form singes, which may 
have been a scribal error for singest, is the only East Anglian 
point of grammar, and naner of sound. 

Norfolk Pronunciation of the Cuckoo Song. 

(S9m-j iz kam in. Ewl'ak stait'eth, 

LEud'h' s/q, kz^kuu' ! Bak wUit'eth, 

Graau'eth seed, Mer'eb* szq, k«^kuu' ! 

And blaau'eth meed, Kwkuu', kwkuu* ! 

And spri"qth clhe wd ueu. AYeI s/q'est dliEu, kwkuu* ! 

S«*q, kt^kuu' ! !N'ot sees dhEu nsevi ueu). 

Zoom, bleet'eth aftM lam, 

Laauth aft"i kalf kEu, 

2. The Prisoner's Prayer (with the Music), circX a.d. 1270. 

In the Eecord Eoom of the Town Clerk's Office in the Guildhall 
of the City of London, is preserved an old quarto vellum manuscript 
known as the Liher de Antiquis Zegihus, of which a re-arranged 
transcription was made by Mr. Stapleton for the Camden Society,^ 
and a translation has been more recently published by Mr. Eiley.^ 
^Neither of these works mention a poem in JS'orman French and 
English, with musical notes, which is inserted at the end of the 
volume, although Mr. Stapleton gives passages which occur imme- 
diately before and after it, and upon one of the pages of the song. 
Both transcriber and translator seem to have considered the song as 
worthless, or as irrelevant to the other matters in the book. 'No 
doubt it did not form part of the work. It seems to have been in- 
serted as a useful piece of parchment, and the old numbering of the 
folios does not go so far. But it is entirely in a xiii th century 
hand, exactly similar to that of the Cuckoo Song, and the musical 
notes, although not written in strict time, are of precisely similar 
forms. It would seem to be a piece of parchment and writing older 
than many parts of the book itself, and probably coeval with the 
Cuckoo Song.^ The music is adapted to the Erench words, which 

* De Antiquis Legibus Liber. ^ Jienry Thomas Riley, Chronicles 

Cronica Maiorum et Vicecomitum of the Mayors and Sheriffs of London 

Londoniarum et qnedam, que contin- a.d. 1178 to a.d. 1274. London. Triib- 

gebant temporibus illis ab anno ner. 4to. 1863. 

MCLxxYiii ad annum mcclxxiv*^; cum ^ The following notes will enable 

appendice. Nunc prinium typis man- the reader to insert this song correctly 

data curante Thoma Stapleton. 1846. in Stapleton's transcript. The numbers 



are carefully placed under their notes, but the English translation, 
written under the French, is not kept strictly under the correspond- 
ing notes and often runs to a considerable length beyond the French. 
Eoth begin together at the beginnings of stanzas. There are several 
mistakes in the English, and one word deleted in the French and 
not restored. This and the absence of musical notes to the few last 
words, shews that the manuscript was not properly revised. It is 
therefore necessary to add a corrected text (pp. 435, 437), which is 
that followed in the subsequent remarks.^ 

The notes, which are now first published (pp. 432-3), presented 
considerable difficulty, from their being written in plain chant, and 
therefore without any division of time, the length of the notes being 
left to the feeling of the singer, as in modem recitative. In the 
following edition I have duly translated the pitch of each note, and 
expanded the ligatures into slurred notes, placing the French words 

in brackets are those of the folios 
numbered in an ancient hand, the 
other numbering is modern and in 
pencil. I have to thank the courtesy 
of Mr. Town Clerk for allowing me to 
inspect the book and make such ex- 
tracts as were necessary. 
Fo. [157], a. Fuit vir quidam, Stap. 
238. This ends on fo. [158], a, last 
paragraph. This folio contains, Iste 
vero. A. natus fuit anno domini mo . 
ducentesimo primo, Stap. 239. The 
Mem., 1086, Stap. 253, ke la Reync 
Isabel etc. L'an E. xx. is in a totally 
different hand. 
Fo. [159], <?, the six Latin lines, Stap. 

253, In hoc folio continentur etc. 
Fo. 159, b, is blank, but both 159, a 
and b are ruled for double columns 
and for writing. 
Fo. 160, a, is blank and not ruled, ap- 
parently an old piece of parchment, 
used and put in. 
Fo, 160, b, and 161, a, the words and 

music of the Prisoner's Prayer. 
Fo. 161, b, the last words of the same 
Prayer, viz. " ei jor et doint ioye 
certeyne," and "we moten Ey and 
o habben the eche bliffe," without 
either musical notes or staff. This 
page also contains the notice : Cum 
de edifices, Stap. 253. 
Fo. 162, «, the five lines, Una Nero 

die, Stap. 253. 
Fo. 162, b. A hymn consisting of ten 
lines and a half of musical staff, 
with Latin words : In translatione 
beati thome, the whole crossed out 
with one cross. 
Fo. 163, a and b. The notice of Thed- 
mar, Stap. 239, Fo. 163 b, is the last 

written page, there are however 
three other blank folios, and one 
with scribbling upon it, which ends 
the book. 

The handwriting of the Prisoner's 
Prayer corresponds with that in the 
best and oldest writing in the book, 
and cannot be later than 1250. 

^ The English text of the Prisoner's 
Prayer appears to have been first pub- 
lished in the Reliqua? Antiqua) i, 274, 
fi'om a transcription by J. 0. Halliwell, 
which reads, incorrectly, v. 1, nun 
for m i w , V. 1 6 1 i c t h for 1 i c 1 1 i, v. 
26 p r s u n for p r e s u n , v. 38 us for 
h u s and v. 39, m i s s e for m i 1 s e , 
and arranges v. 13, 14 thus 

For othre habbet misnome 
Ben in this prisun i-broct. 
The present copy is re-printed, from 
the work cited below, p. 498, n. 1, with 
an improved stanza III, and the cor- 
rection V. 41 wu fit go for wn fit. 
go, the result of renewed inspection. 
The corrected text has also been re-cor- 
rected, especially in the verse last cited, 
where Dr. Stratmann's conjectui'e that 
go wu fit go stands for go hu so 
it go has been adopted, wu=whu, 
being a not unfrequent form of h u in 
the XIII th centmy, (infra p. 440,) 
and the contraction sit for so it 
being partially justified by Orrmin's 
^ho't for ^ho itt = she it, and 
h e' t for he i 1 1. Most of the other 
corrections are evident enough. The 
only difficult word i p e 1 1 is illus- 
trated below, p. 448. See also : ]?at 
wer for sin in belle ipilt ; of paradis 
hi wer ute pilt ; fort godes soue in rode 



Chap. V. 

under the notes as indicated in the original.^ But I have taken the 
liberty of reducing the time to a modem system, and have added 
bars accordingly.''^ As frequently happens in translations, the Eng- 
lish words do not in all cases exactly correspond to the notes written 
for the French. This has occasioned much difficulty in adjusting 
the corrected text of the English words to the notes, and such 
changes in the music as have appeared necessary are indicated by 
smaller notes. When two sets of notes appear in one bar, the 
direction of their tails shews in the usual way to what version they 
refer. It is evident that no stress can be laid on any passages in 
which such alterations have appeared necessary, as regards the 
pronunciation of the syllables.^ Enough passages remain in which 
final -e was undoubtedly pronounced, to establish here as well as in 
the Cuckoo Song, the general rule for pronouncing it. At the same 

prisun in y. 4, and the word Christ in 
V. 7, have each in the MS. two identi- 
cal repeated notes written close together. 
In each case I have reduced these to 
a single note, as I have been unable to 
obtain any explanation of this doubling. 

2 The key is the ecclesiastical mode 
of which the scale ran from G, thus G 
A B c d e f g, without any sharps or 
flats. Each stanza is treated as a sepa- 
rate composition, and the second half 
of each stanza repeats the music of the 
first half, almost precisely. This has 
enabled me to supply the missing notes 
of the fifth stanza, answering to the 
French words : " ^^ jor et doint ioye 
certeyne," with almost perfect cer- 
tainty. I am indebted to Mr. "Wm. 
Chappell for much information respect- 
ing the meaning of the old musical no- 
tation, and for an acquaintance with 
the important works of E. de Cousse- 
maker : (Scriptorum de Musica Medii 
-^vi novam seriem, 1864, 4to., and 
L'Art Harmonique aux xii et xiii^s 
siecles, 1865, 4to.) without which I 
could not have translated the music at 
all. But for the barring of the Pri- 
soner's Prayer, I alone am respon- 
sible, and I have been guided entirely 
by the symmetry of the musical pas- 
sages and the rhythm of the words, 
not at all by any possible indications of 
length in the notes themselves, as was 
the case in the Cuckoo Song, in which 
the time is accurately indicated. 

3 Thus we cannot be quite sure that 
the singer pronounced shame v. 4 in 
two syllables, although there seems to 
be no doubt that he said name v. 5 in 
two syllables. Similarly so7ne, misnome, 
V. 11, 13, may have omitted the final 
-e for the music. 

was pilt, Furnivall's Early English 
Poems, p. 13, V. 8 and 35 ; p. 14, v. 
56, from Harl. MS. 913. The French 
text has been printed by M. Jules Del- 
pit, in his Collection Generale des Docu- 
ments FrauQais qui se trouvent en 
Angleterre, Paris, 1847, 4to. vol. 1, 
p. 28, No. LXVII. This transcript is 
faulty having d'anguste for dangusse 
V. 2, dur for d u z v. 6, en suit for 
enset v. 12, U sires Deus ke for 
Sire deus ky v. 15, I eel for 
i c e r V. 23, morteu for m o r t e n 28, 
fort for fors v. 30, guee for guie 
V. 34. The u and v are also modern- 
ized, the stanzas not divided as in the 
original, some contractions expanded 
without notice and others not, the 
omission of et v. 39 not perceived and 
V. 5 made to end with tres puis instead 
of Ihesu, in defiance of the metrical 
point, the metre and music. In citing 
the J^el. Ant. for the English version, 
M. Delpit prints Hallewell, Shraps, 
Pikering for HalUwell, Scraps, Pic- 
kering. He says of this poem (ib. p. 
cxcii) : " Le No lxvii est le plus 
ancien document en vers publie dans 
ce volume. Je I'ai trouve sur les 
feuillets de garde d'un manuscnt du 
xiii^ siecle, connu dans les archives de 
la raairie de Londres sous le nom de 
Liber de antiquis legibiis ; mais sa com- 
position pent remonter a une epoque 
beaucoup plus ancienne que celle de sa 
transcription .... il m'a paru important 
par son anciennete, et de nature a four- 
nir quelques remarques utiles sur les 
regies qui presiderent a la formation 
de la langue que nous parlous." 

^ In three instances only have I 
deviated from the original. The se- 
cond syllable of pleynte in v. 1, and of 

§ 1, No. 2. prisoner's prayer XIII TH CENTURY. 431 

time other passages occur in which it seems to have been un- 
doubtedly omitted, not only before a vowel, but elsewhere, and 
these are all indicated by an apostrophe in the corrected text.^ 

The rhymes are generally quite regular, but there are a few 
anomalies which prepare us to look out for assonances intermixed 
with perfect rhymes in poems of the xinth century and earlier. 
Thus: man am 7, 9 ; hem men 21, 22 ; live bilive sti^e 27, 28, 
29 ; mildse blisse 39, 44 ; are all assonances (p. 245, note). But 
they are assonances which many ears mistake for rhymes, because 
the differences of the consonants are not obstrusive. The French 
version has also the assonance : deus mortels, 15, 16 ; and perhaps : 
euayn heim, 37, 38. 

As regards the orthography in the uncorrected text, the use of d 
for ^ is common enough in other MSS. not to need explanation ; 
the he for ch is an occasional carelessness, compare ihe 4, with ich 
1, 2, 3, found also in the Proclamation of Henry III. ; and the 
occasional insertion of h is frequent in Layamon, and may indicate 
a doubtful pronunciation, compare vs 20, with hus 40, 41. More 
noticeable is the invariable use of th for \ at so early a period, and 
gh or occasionally yh (forghef 21, yhef 23) for 5 ; the use of ct for 
5^ (noct 12, ibroct 14) is not otherwise uncommon. The orthography 
yh seems to point to a {fii) or (jh) as preceding the use of (j), where 
5 occured in ags., as already suggested (p. 313). Wos 24 for whos, 
and, if Dr. Stratmann is correct, wu 42 for whu and that for Aw, may 
be assimilated to the cases of inserted h, as shewing a lack of appre- 
ciation of the aspirate. The use of c for 5 in such words as hlisce 
31, 44, is not uncommon, compare Gen. and Ex. 3518. 3fai 28, 
for the older form ma"^, and maiden 35, indicate that the diphthong 
had been completely formed from a-^ (ag, agh, a^^h, ajh, ai) ; and 
ey 43, compared with Orrmin's ^55, shews that a wi^iter did not 
feel any difference between the diphthongs (ei, ai), which Sir 
Thomas Smith found it so hard to distinguish three centuries later 
(p. 121) and which were constantly confused in the xivth century 
(p. 263). These are the only words in the English text bearing on 
these diphthongs. But in the Erench we have, souerein, mayn, 
euayn, heim 35, 36, 37, 38, rhyming together, and we have plest, 
forfet 24, 25, indicating an unpronounced s before t, and a degene- 
ration of ai in certain words into (e) even at this early period. 
The Prisoner's Prayer never uses ou for (uu), but employs u as 
in Tcuthe 1, nu 2, thu 8, prisun 9, ut 10, huten 34. The sume 11, 
and misnome 13, are either errors for sume, misnume, or some, mis- 
name, probably the latter, as same, some are the ags. forms. There 
is no instance of u being employed for i, e or ags. y. The Erench 
text, to which the notes were primarily adapted, raises the question 
of the pronunciation of N^orman. See p. 438. 

1 Final -e, ehded before a vowel, a consonant, )>in' 5, hop' 27, bar' 35, 

ku]?' 1, sor' 3, bal' hal' 17, wel' 31, son' 36, liv' 42; internal e omitted, 

but' 34 (this is a conjectural emenda- much'le 4, hev'ne 18, 35 ; and if Dr. 

tion), habb' 37, bring' 40 ; before an Stratmann' s correction is adopted we 

H, ojjr' habbej? 13, rajj' he 32 ; before have s'it for so it, v. 42. 



Chap. V. 


From the Liber de Antiquis Legibus^fo. 160 b. 

Note. The French as in the Original MS., the English according to the Corrected Text. The 
slurred and joined notes represent the original ligatures. The time and bars ai'e modem, the 
original being in plain chant. The last five bars are not in the MS., but have been supplied 
from the parallel passage commencing with the bar marked *. 

\_Adagio, affettuoso.'] 


J ( Eyns ne soy ke pleyn 
' ( Ar ne ku}j' ich sor ■ 

- te fu 
^e non. 

- re pleyn dan - gus - se tres 
Nu ich mot ma - nen min 





-» — •- 


■^9-^\m — 0- 






su trop ai mal et con-trey-re Sanz de - car - te ^n pri-sun sui. carmay-dez tres- 
mon. Kar-ful welsor' ich si-che. Giltles ich tho-liemuch-lescha-me Help God for thin 



pu-is Ihe-STj. diiz deus etde-bon-ney-re.jy | Ihe-sucristveirsdeuueirshom.preng-e 
svfe-te na - me, King of hev-en - e ri-che. ' \ Je-su Crist,*sojj God,*soJj man, Lhoverd, 






- ^=^ ^ -p- -•— P= '■ 

vus de mei pi - te. Je-tez mei de la pri-sun v ie sui a-tort ge-te. lo e 
rew pn. up -on me ! Of pri-sun j^ar-in ich am Bring me ut andmak-ie fre! Ich and 

mi au-tre com-paign-un deus en-set la ue-ri - te. tut purau-tremes-pri-sunsu-mes 
mi - ne fe-ren so-me (Godwotjichneli^-eno^t,) ForoJ7r'habbe)7benniis-no-me[And]in 


a hun-te li - ue - re. 
this pri - sun i - bro^t. 

jjj ( Sire deus ky as mor-tels es de par - dun 
^^^' [ Al-mi^ - ti )7at wel li3t - li Of bal' is hal' 




ue - i - ne. su - cu - rez de - K-ue-rez nus de ces 

and bo - te. Hev'-neking! Of Jjis won-ing Ut us brin 

te pei 
gen mo 

§ I, No. 2. prisoner's prayer XIII TH CENTURY. 








ne Par-don - ez. et as - soy - lez. i - eel' gen - til si - re. 

te, For - :^ef hem J?e wik - ke men God ^if it is J^i wil - le. 





_i — -F-x 


-• — •- 

Si te plest par ki for - fet 
For whos gilt We beoj? i - pilt 

nous suf-frun tel mar - ti - re. 
In jjis pri-sun il - le. 





-9 — •- 


( Fous est ke se a - fi - e en cest-e mort-cn 
|Ne hop' non to his live! Heme ma^ he hi 

u - ie. ke tant nus con-tra- 
li-ve He - ^e j^e^ he 


li - e Et V nad fors boy,- di - e. Ore est hoem en le - es - se et ore est 
sti - ^e de)? fel-lej? him to grunde. Nu ha|7 man wel' and blis - se, Raj?' he schal 







en tris - tes-ce ore le ga - rist ore bles-ce for - tu-ne ke le gui-e. 

jjar - of mis-se World-es we-le, mid i - wis-se Ne las-te|j but' on stunde. 

XCon Forza.l 



-0 — 0- 

— I- 



Y ( Vir - gine. et mere au so - ue - rein, ke nus ie - ta de la ma - yn Al mau-fe 
[ Ma^-den |?at bar' \}e hev'-ne king, Bi-sech \}m son', j^at swe-te J?ing, )?at he habb' 






ki par e - uayn nus ont tres-tuz en sun heira a grant do-lur (et) pein-e. 
of us raw - sing And bring' us ut of )?is wo - ning For his mu-chel-emild-se 







Re - que-rez i - eel sei - gnur ke il par sa grant dul-cur nus get de ces-te 
He bring' us ut of j^is wo, And us te - che werchen swo In f is liv' go hu 







do - lur. V nus su-mus nuyt et 
s'it go, jjat we mo-ten a^ and 

Jor et doint ioy - e cer - tey - ne. 
0, Hab-ben ]pe ech-e blis-se. 




Chap. V. 

From tlie Liher de Antiquis LegiluSy Guildhall, London, fol. 160 J. 

Norman French Original. 


Eynf ne foy ke pleynte fu 
ore pleyn dangufle treffu 

trop ai mal et contreyre 
Sanz decerte en prifun fui. 
car maydez trefpuif ih^u. 

duz deuf et deboneyre. 


Ih^u crift veirf den ueirf horn. 

prenge vuf de mei pite. 
Jetez mei de la prisun 

V le fui atort gate. 
lo e mi autre cowpaignun 

deu5 enfet la uerite. 
tut pur autr^ mefprifuu 

fumes a hunte liuere. 



Sire deuf 
ky af mortels 

ef de pardun ueine. 17 


nuf de cefle peine. 20 

Early English Translation. 


Ar ne kuthe ich forglie non. 
nu ich mot manew miw mo^. 

karful welfore ich fyche. 
Geltles ihc sholye muchele fchame 
help god for thiw fwete name 

kywg of heuene riche. 


Jefu crift fod god fod man 

louerd thu rew vponme 
of pnfun thar ich m am 

briwg me vt and makye fre. 
Jch and mine feren fume 

god wot ich ne lyghe noct 
for othre habbet mif nome ben 

in thyf pr^fuw ibroct. 


Al micti 
that wel lictli 

of bale if hale and bote, 
heuene king 
of this ■woniwg 

Tt vs briwge mote. 

Verbal Translation of the Norman French. — I. Once (I) knew not what 
affliction was, Now, full of anguish, tormented {fres sue)^ Too much (I) have (of) 
ill and misfortune. "Without guilt in prison am (I), Wherefore help me right 
soon {fres puis) Jesus, Sweet God and gracious. II. Jesus Christ, true God, true 
man. Take you pity on me. Cast me from the prison, Where I am wrongfully 
thrown. I and ray other companion, God knows of it {en sail) the truth, All for 
other mistake (in mistake for others), Are delivered to shame. — III. Sire God, 
"Who to {aux) mortals Art of pardon source \veine)y Help, Deliver Us from this 

J ^ 



Corrected Text 


Ar ne ku]?' icli sorge non, 
I?'u ich mot manen min mon. 

Karful wel sor' ich siche. 3 
Giltles, ich folie much'le schame. 

Help, God, for fin' swete name, 

King of hevene riche. 


Josu Crist, soj? God, so]? man, 

Lhoverd, rew f u upon me ! 
Of prisun ]7arin ich am, 

Ering me ut and makie fre ! 10 
Ich and mine feren some, 

(God wot, ich ne lige nogt,) 

For ofr' hahbef hen misnome 

[And] in Jis prisun ibrojt. 14 


pat wel lijtii 

Of bar ishaP and bote, 17 
Hev'ne king ! 
Of fis woning 

Ut us bringen mote. 20 

Conjectured Pronunciation. 


Aar ne kuuth itsh sor*ghe noon, 
Nuu itsh moot maa-nen miin 
Kaar'ful* wel soor itsh siitsh*e. 
Gilt-les, itsh thooiie mutsh-le 

Help, God, for dhiin swee'te 
Kiq of Hee'vene riitsh-e. 


Dzhee'su Krist, sooth God, sooth 

Lhoverd, reu dhu upon* mee ! 
Of priisuun" dhaarin* itsh am, 

Eriq me uut and maa-kie free ! 
Itsh and mii-ne fee'ren soo -me, 
(God wot, itsh ne lii-^he 
For oo'dhr- -ab'eth been mis- 

noo 'me 
[And] in dhis priisuun* ibrokht*. 


Dhat wel li7{;ht-lii* 

Of baal is Haal and boo-te, 
Heevne king ! 
Of dhis woo'niq- 

Uut us briq-en moo'te. 

Verbal Translation of the Early English {corrected text). — I. Erst not knew I 
sorrow none, Now I must moan (ags. mcenan) my moan. Ful of care right 
sorely I sigh. Guiltless, I suffer much shame. Help, God, for thy sweet name, 
King of heaven's kingdom. — II. Jesus Christ, true God, true man, Lord, rue 
thou (have mercy) upon me ! Of (the) prison wherein I am, Bring me out and 
make (me) free ! I and my companions (plural here, singular in the French) 
together (God knows, I not lie nought), Have been for others mistaken, i.e. 
wrongfully taken, [And] in (to) this prison brought. — III. Almighty, That weU 
easily Of harm is healing and remedy. Heaven's king, Of this affliction May (he) 
bring us out. 


prisoner's prayer — XIII TH CENTURY. Chap. V. 

Norman French, 
et aflbylez. 

iceF gentil fire. 23 

SI te pleft 
par ki forfet 

nuf fuffruw tel martire. 26 


Fouf eft ke fe afie 
en cefte morten me. 
ke tant nuf contralie. 

Et V nad fors boydie. 30 
Ore eft hoem en leeffe 
et ore eft en triftefce 
ore le garift ore blefce 

fortune ke le guie. 34 


Virgme. et mere au fouerein. 
ke nuf leta de la mayn 
al maufe ki par euayn 
nuf ont treftuz en fun heim 

a grant dolur [et] peine. 39 
Requerez icel feignwr 
ke il par fa grant dulcur 
nuf get de cefte dolur. 
V nuf fumus nuyt et Jor 

et domt loye certeyne. 44 

JEarly English, 
Foryhef hem 
the wykke men 

god yhef it if thi will© 
for wof gelt 
we bed ipelt 

in thof pnfun hille. 


"Eq hope non to hif liue 
her ne mai he biliue 
heghe thegh he stighe 

ded him felled to grunde. 
iN'u had man wele and blifce 
rathe he ftial thar of miffe. 
woiides wele midywifle 

ne lasted buten on ftunde. 

Maiden that bare the heuen king 
bifech thin fone that fwete thing 
that he habbe of hus rewfing 
and bring hus of this woniwg 

for his Muchele milfe. 
He bring hus vt of this wo 
and huf tache werchen fwo 
m thof Hue go wu fit go. 
that we moten ey and o 

habben the eche blifce. 

Verbal Translation of the Norman French, continued. — Pardon And absolve 
Him, gentle sire, If (it) thee please, By whose crime We suffer such martyrdom. 
— IV. Mad is (he) that has confidence In this death in life {mort en vie,) Which 
afflicts {contralie = contrarie, Roquefort) us so much, And where (there) is nothing 
but deceit {et ou n*a=i\ n'y a, Aors=que, Jo?/<?i<9 = boisdie— voisdie, from 
versutia). Now is man in joy, And now (he) is in sorrow. Now him heals 
{guerit), now wounds, Fortune who guides {guide) him. — ^V. Virgin and mother 
to the sovereign Who cast us with his {la, lit. the as in modern French) hand To 
the devils {aux malfaits), who through Eve {Evain) Have us right all {tres tons) 
on their hook {heim, haim, 7iam = Latin hamus, modern hamegon) In great grief 
and (supply et, wanted for the construction, metre, and music, the word originally 
written has been erased,) pain. Beseech that Lord, That he by his great sweet- 
ness {douceur) May cast us from this grief, Where we are night and day, And 
give {donne) sure joy. 



Corrected Text, 
Forgef hem 
J7e wikke men, 

God, gif it is J?i wille, 23 
For whos gilt 
We beo]? ipilt 

In J7is prisun ille. 26 


Ne hop' non to his live ! 
Her ne mag he bilive. 
Hege feg he stige, 

Def felle]? him to grunde. 30 
!Nu ha]? man wel' and blisse, 
Ea]?' he schal farof misse. 
"Worldes wele, mid iwisse, 

iNe lastej? but' on stunde. 34 


^Magden, fat bar' ]?e hev'ne king, 

Eisech fin son', fat swete f ing, 

pat he habb' of us rewsing, 
And bring' us of this woning. 

For his muchele mildse. 39 
He bring' us ut of this wo, 
And us tache werchen swo, 
In f is liv' go hu s' it go, 
pat we moten, 35 and o, 

Habben f e eche blisse. 44 

Conjectured Pronunciation. 
Forjeef* Hem 
Dhe wik'e men, 

God, jif it is dhii wil'e. 
For whoos gilt 
"We beeuth ipilt* 

In dhis prii'suun il-e. 


ITe Hoop noon too nis lii-ve ! 
Heer nee mai nee bilii-ve. 
Hekh-e dheekh ne stii-^he, 

Deeth fel'eth nim to grund"e. 
N'uu Hath man weel and blis-e 
Raath ee shal dhaar-of mis*e. 
"World-es weel'e, mid iwis'e, 

l^Q last'eth buut oon stund-e 


Maid-en dhat baar dhe neevne 

Biseetsh- dhiin soon dhat sweet-e 

Dhat He nab of us reusiq*. 
And briq us of dhis woo'niq* 

For His mutsh*el*e mil-se. 
Hee briq us uut of dhis woo 
And us taatsh'e wertsh'en swoo. 
In dhis liiv goo huu s- it goo, 
Dhat we moo -ten, ai and 00, 

Hab-en dhe eetsh*e blis*e. 

Verbal Translation of the Early English {corrected text), continued. — Forgive 
them The wicked men, God, if it is thy will, For whose guilt "We (have) been 
thrust In (to) this vile prison. — IV. Let none have trust in his life! Here 
may he not remain. High though he rise, Death fells him to (the) ground. 
Now hath one weal and bliss, Suddenly he shall miss thereof. (The) world's weal, 
with certainty, Lasteth not but one hour. — V. Maiden, that bare the heaven's 
king, Beseech thy son, that sweet thing, That he have of us pity, And bring us 
out of this affliction, For his great mercy. May he bring us out of this woe, 
And so to act teach us, In this life go how so it go, That we may, aye and ever 
Have the eternal bhss. 


An examination of the pronunciation of old Prench, especially of 
the T^orman dialect, is also almost forced upon our attention by 
the close connection of the two languages during the formation of 
English proper. The researches now being instituted by Mr. J. 
Payne into the persistence of Norman forms ^ have given the 
pronunciation of I^orman a still greater interest. The investigation 
is fraught with difficulty, as will appear at once from the present 
attempt to resuscitate early English sounds. It must be conducted 
separately, first by an examination of all the documents tending to 
throw a light upon early French pronunciation ; secondly, by a careful 
study of the living dialectic pronunciation in the IS^orth of Erance ; 
thirdly, by a review of Korman French poetry, either in original ma- 
nuscripts of known dates or in trustworthy editions of the same, 
such as M. Michel's edition of Benoit.'^ To assume that old Norman 
was pronounced as modem Norman,^ or modern French, or modem 
English, would be against all historical precedent, and the most 
probable hypothesis is that it differed from all of these in many 
respects, but that we may find indications of the existence of all of 
the latter forms in particular cases. Such an investigation is 
entirely beside the present, although both have been occasionally 
brought in contact, through Palsgrave in the xvi th century, and 
such translations from the Norman as the Prisoner's Prayer, and 
the rhymes of English and French in Chaucer and the Political 
Songs. It would be difficult for any but a Frenchman to conduct, 

^ " The Norman element in th^ speech, says : " On ne pent, h mon 

English, spoken and written, of the avis, generaliser aucune assertion sur 

XIII th and xiv th centuries, and in the les points de detail, attendu que I'ex- 

provincial dialects," is the more ex- pression et meme I'accentse localisent 

tended title which Mr. Payne has eitremement .... Ce qui est vrai ici, 

adopted for his papers read before the peut ne pas I'etre la. . . . Chez nous 

• Philological Society in 1868 and 1869. (dans le diocese de Rouen) on trouve 

2 Chronique des Dues de Normandie deux dialectes completement differents 
par Benoit, trouvere anglo-normand du d' accent : le brayon, parle dans la 
XII ® siecle, publiee pour la premiere portion orientale du departement (ou 
fois d'apres un manuscrit du Musee diocese) surtout dans I'arrondissement 
Britannique par Francisqiie Michel, de Neufchatel, et une portion de celui 
1836-1844. 3 vols. 4to. Published by de Dieppe. L'accent est picard, par 
order of the French government. The consequent href, et avec le systeme de 
MS. followed is Harl. 1717, and the syncopes propres au picard: e' veuV- 
printed text was compared with the tent Men, mats i' n' peuv'tenf pas, 
original by Sir F. Madden. There is ils veulent bien, mais ils ne peuvent 
a copy in the Reading Room of the pas. Du reste pas de mots originaux. 
British Museum. Le cauchois, parle dans tout le plateau 

3 It would be as wrong to suppose occidental allonge extremement la der- 
that there is a Norman dialect, as that niereoul'avant dernieresyllabe dumot, 
there is a Scottish dialect. Both of prononcel'atresouvert: le dialecte cau- 
them admit of separation into several chois est riche en mots originaux, mais 
distinct forms, requiring different forms ces mots sent fort localises." The "has 
of writing to be intelligible. M. I'abb^ Normand." speaks, again, a different 
Delalonde, professor of history at the set of dialects. Hence, although we 
faculty of theology at Rouen, who has may find remnants of old pronunciation 
most kindly replied in writing to in all these dialects, it would be hazard- 
several questions which I took the ous to infer the old pronunciation from 
liberty of putting to him on Norman any one of them. 


and we may probably have to wait for a considerable time, before a 
properly qualified investigator devotes himself to the task. May 
this last anticipation prove incorrect ! ^ 

3. Miscellanies of the xiii th Century from Reliquije Antiques;, Early 
English Poems, and Political Songs, with an Examination op the 
Norman French EI, AI, 

Under this heading some brief notices will be given of short 
rhymed pieces belonging to the xni th or the earliest part of the 
XIV th century, contained in the Reliqum Antiquum ^"^ Early Eng- 
lish Poems,"^ and Political Songs} 

The most considerable poem in the Peliquice AntiqiKB is the 
Bestiary, i, 208 ; ^ it is only partly in rhyme, ^ and the rhymes are 
not unfrequently broken by non-rhyming couplets, or fall into mere 
assonances, so that no reliance is to be placed upon them for deter- 
mining the pronunciation. Thus we cannot be sure that 5, which 
is used throughout the poem for sA, was pronounced (s), from the 
rhyme : fis is, p. 220, v. 499, 529, for between them we have : 
biswiken bigripen, v. 515. Other parts are alliterative and there- 
fore of no assistance, but they burst out occasionally in rhyme for a 
few lines. This poem uses v, consistently for (uu), and ow, ow for 
(oou, ou) as in : out p. 223, v. 645 == aught, nout p. 209, v. 18 = 
nought, occasionally written nogt, p. 212, v. 187, sowles p. 211, 
V. 118, soule p. 213, v. 206, knowe^ p. 211, v. 121, knov p. 212, 
V. 165. There seems to be no use of u for i or e thi'oughout the 
poem, thus we have : mirie p. 221, v. 570, pit p. 226, v. 761 ; this 
consorts properly with the consistent use of u for (uu). Similarly 

^ Diez, Grammatik der romanischen British Museum by F. J. Furnivall, 

Sprachen, 2nd ed. 1856, vol. 1, pp. 1862, for the Philological Society. 

404-454, investigates the meaning of * The Political Songs of England 

the old French letters, but leaves much from the reign of John to that of 

to be desired. The commencement of Edward II, edited and translated by 

an investigation into the values of Thomas "Wright, Esq. London, 1839. 

Norman a, ai, together with a few 4to. pp. xviii, 408. Camden Society, 

other casual remarks on old and modern ^ The text of this was especially read 

Norman pronunciation, will be found by the MS. Arundel 292, fo. 4. for the 

below, p. 453. See also the extracts Eel. Ant. It has been reprinted with 

from iDr. Eapp, below, p. 509, n. 1. extensive notes, and a few conjectural 

„ T^ , . . . , . CA /. emendations, in : Altenglische Sprach- 

2 Rehquise Antiquae. Scraps from ^^^^^ ^^^^^ ^^^^^ Woi terbuche, unter 
Ancient Manuscripts illustratmg chiefly foit^irkung von Karl Goldbeck heraus- 
Early English Literature and the gegeben von Eduard Matzner. BerUn, 
English Languge. Edited by Thomas f g^y ^ 8vo vol i p 57 
Wright and James Orchard Halliwell, « rp^ followins- narts are in rhvme 
?;,t- ^r- PP;\27 -d 287 1841- ^^^ p\ges ref" fo'the S T^'The 
1843 The text has generally been n^^? of the lines are taken from 
carefully transcribed and printed but Qoi^beck and Matzner : p. 209, v. 40- 
some mistakes occur as pointed out ^^ 2IO, v. 54-87 in couplets v. 89- 

P- ..^ T ®o ' P- ^ ' ''''*' ' 119 alternate rhyme, p. 215, v. 307 to 

p. 445, note L. ^ ^ 217, v. 384, with a few short inter- 

3 Early English Poems and Lives of ruptions, p. 218, v. 424 to p. 219, v. 
Saints (with those of the Wicked Birds 455, p. 220, v. 499 to p. 221, v. 554, 
Pilate and Judas). Copied and edited p. 222, v. 588, to p. 224, v. 694, p. 225, 
from manuscripts in the Library of the v. 733 to p. 227 v. 802 and last. 


we rarely find o for either (uu) or (u), thus : sunne = sun, p. 209, 
V. 19, cunne p. 210, v. 69, come p. 209, v. 35, but: cume^ p. 210 
V. 67, sumer p. 214, v. 236, hule p. 214, v. 253, but : hole p. 217 
V. 394, p. 226 V. 769, and the rhyme : cul ful p. 225 v. 741 = 
cowl full,^ may be considered as establishing the value of long u as 
(uu) in opposition to the modem opinion that it is (ou) or (ou). 

The spelling is generally good and consistent,'* but it presents 
certain peculiarities. Thus s is always employed both for s and sA, 
and the rhyme, as akeady pointed out, ought to deteimine that (s) 
was the only sound. Also g is used throughout, generally as pure 
5 with the guttural effect after vowels, as in : sigte p. 211, v. 107, 
rigten p. 211, v. 117, drigtin p. 211, v. 119, ^urg p. 211, v. 119, 
inog p. 211, V. 142. Sometimes the resolution into (j) or (i) seems 
indicated by a prefixed «", as: leige^ p. 216, v. 359, maig p. 210, 
V. 80, p. 220, V. 516, p. 221, v. 548, but the ^ is then most generally 
omitted as in: mai p. 211, v. 129, mainles = without power, main 
force? p. 211, v. 128, dai p. 210, v. 63, but dei p. 215, v. 305, 
meiden p. 209, v. 37, shewing that ai, ei were confused. Initially 
the g was simply (j) to judge by: ging = young, p. 213, v. 214, 
gu = you p. 244, v. 700, ge = she p. 214, v. 243, but it may have 
been (^h). After i it disappears altogether as: sti p. 213, v. 198. 
The aspirate h is treated very irregularly, being sometimes start- 
lingly inserted, as hac for ac p. 226, v. 792, and frequently omitted. 
After w it generally disappears, as : wit, wel = white, wheel, 
p. 225, V. 737. The form wu for whu =^'hu = how (supra p. 429, 
note 1,) is frequent, as p. 209, v. 36 and v. 55, but: hu p. 210, 
V. 56, in the next line. The pronunciation of ch seems intended 
for (tsh), and such apparent rhymes as : riche ilike = rich alike 
p. 222, V. 604, must be considered as assonances, unless we suppose 
-like to be an orthographical error for -liche. The use of ^ is 
general, but we have bicumeth p. 210, v. 91, unless it be a mis- 
print. After s, t, d this ^ becomes t, as in Ormin, the instances 
are collected by Matzner at v. 22. 

The diphthongs at, ei appear to be (ai) by the cases already cited. 
Forbroiden p. 211, v. 124, seems to stand for forhrogden and should 
imply therefore oi = (oi), but it is uncertain, and similar oi diph- 
thongs are unknown, so that we cannot infer generally oi = (oi). 
In : newe p. 225, v. 724, spewed p. 211, v. 139, ^ewes p. 212, v. 
183, reufulike p. 223, v. 652, we can hardly take eu for anything 
but (eu). In : taunede p. 226, v. 767, middle high German zounen 
to shew, (au) seems to be implied. 

1 "Wor so lie "wiine'S ^is panter, diversis pastus venatibus. The ags. 
he fede'S him al mid o'Ser der, cufle {JElf. gl. 20), cugle (i^.), cuhle, 
of "So "Se he wile he nime^S "Se cul cowl is remarkable for the early in- 
and fet him wel til he is ful. terchange of (f, gh) which has not 
='WTiereso he dwelleth, this panther, descended. If cul is to be thus inter- 
he feedeth him all with other deer preted, it has lost a final e. But is not 
(beasts), of those that he wil, he taketh rather cul the French word meaning 
the cowl (skin ?) and feedeth him wel rump, the prime piece ? 
til he is full. This is Matzner' s inter- 2 xhe handwriting of the MS. is 
pretation of cul. The Latin has only : particularly beautiful, large, and careful. 


On the whole this poem, though presenting some peculiarities, fully 
confirms the conclusions derived from the two preceding old poems. 
In none of the others does the orthography seem so trustworthy. 

The Family Prayer, Paternoster, etc., vol. i, p. 22, mixes as- 
sonances with its rhymes freely, as : lif siche, bunden wndes ; king- 
dom don ; wndis bunde. Of these : lif siche = sickness, is useful in 
establishing the value of the long i as (ii) or («). The u is consis- 
tently used as (uu), and ou in trotie as (oou), once erroneous spelled 
true, but au is also used in saulc, which, if correct, is an early and quite 
unusual transformation of suk. The rhyme to this word : bysuak 
seems to imply some error in the MS., which is here correctly tran- 
scribed. Another unusual form is : leyse for lese, and fleyes for 
flesh, compare supra p. 265, and infra p. 473, n. 4. Although Marie 
occurs fully in : Hcil, Marie, ful of grace I = (Hail Marii'e ful of 
graa'se !) it is abbreviated to Mari, in 

Moder of railce/ and maidin Mari, (]\Ioo-dcr of mils, and maidin Marii*, 

Help us at urc bending, for y\ merci. Help us at uur cnd-iq for dhii mersii*.) 

]!^o doubt this was a very ancient occasional abbreviation of a 
name so common on the lips of all worshippers : thus in Germany 
(Maarii*) is fully as common as (Marii'e) in addressing persons of 
that name. See p. 446, Ex. 3. The aspirate comes in curiously in : 
hart = art, hus = us, as well as house, bending = ending, herdejje 
= earthe, hure = our. The guttural is evidently expressed by ch 
in : jjich,^ halmichtcnde, licht, riclit, which is very unusual. 

The Creed and Paternoster, vol. i, p. 57, are not in the pure 
xin th century orthography. We have indeed : ure, wi]7utcn, but : 
Pounce (written Punce = Pontius, in the last example), ous, foule. 
This shews a period of transition, which will be especially noticed 
inHavclok, infra p. 471, occasioned by the growing use of u as (yy) 
or (ii, i, e), compare in the Creed : y-buriid, and in the Paternoster : 
als we forgivet uch ofir man. Other peculiarities here are : sshipper 
= schipper, ags. scyppan, create ; and : fleiss = fleisch, flesh ; steich 
= steg, ascended. The rhymes in the Paternoster are correct, 
except : don man. 

Another Creed, Paternoster, Ave, etc., are given in vol. i, p. 
234, in which the u long is perfectly preserved for (uu), and: 
biriedd, iche, are used. Pontius appears as Ponce, which compared 
with the first Punce, shews the use of o for short (u). The Pater- 
noster is chiefly in assonances, and we cannot feel sure that : deadd 
so^fastheedd, in the next prayer, is a rhyme or an assonance, that 
is, whether the first word is (deed) or (deeth), or (deead). The last 
little moral has some assonances : 

If man him bi^octe (If man -im bitbokb-te, 

Inderlike and ofte In-erliik and oft-e, 

Wu arde is te fore • Huu Hard is te foo-re 

Fro bedde te flore, Fro bed-e te floo-re, 

^ Tbis is the MS. reading, tbe compare \eagh in a sermon of tbe 

printed text \\.di?, milte, ags. mildse, see xiiitb century, from MS. Trin. Coll. 

supra p. 429, note 1. Cam. B. 14, 52, in Eel. Ant. i, 129, 

.. 2 Imperative of ^eon to prosper, 1. 2 and 14. 



Chap. V. 

"Wu reuful is te flitte Huu reu-ful is te flit-e 

Fro flore te pitte, Fro floo-re te pit-e, 

Fro pitte to pine Fro pit-e te pii-ne, 

"Sat neure sal fine Dhat never shal fii-ne, 

I wene non sinne li wee-ne noon sin-e 

Sulde his herte winnen. Shuuld -is Hcrt win-en.) 

Eut we miglit suppose that (bithof'te) was already occasionally 
pronounced, as in the West of England (supra p. 212). The French 
jine^ finir, end, establishes the pronunciation of fine. Fore for fare 
is a North- countryism, and te for the usual to^ seems to indicate an 
indistinct utterance, perhaps (t^). I have ventured to pronounce : 
sal, sulde, with (sh), but I do not feel quite certain, for reasons 
named above, p. 440. 

Immediately preceding this moral is the following in which : I 
ne, occurs in Mr. Wright's text, but : ine, in one word, occurs in 
the MS, just as in the old high German quoted by Graff, (supra 
p. 292, n. 2), and clearly shewing the (in'e) or (ii'ne) pronunciation. 

Wanne I 'Senke 'Singes 'Sre, (Whan i theqke thiq-es three, 

Ne mai hi neure bli'Se ben ; Ne mai i never bliidh-e bee ; 

^e ton is dat I sal awei, Dhet-oon is dhat i shal awai', 

^e to^er is ine wot wilk dei Dhet-oodh-r is in-e wot whilk dai, 

"Se "Sridde is mi moste kare, Dhe thrid-e, is mi most-e kaa-re, 

Ine wot wider I sal faren. In-e wot whidh-er i shal faa-re.) 

In this pronunciation I have taken some necessary liberties with 
the text, as the omission of an Infinitive n for the rhyme, rectifi- 
cation of the aspirate, w for wli^ ^ for ^, etc. 

The three first Paternosters, Aves, and Credos, are here given for 
comparison with those of Dan Michel, supra p. 413. They have 
been read with the original MSS.,^ and are printed accordingly, 
with the exception of capitals, punctuation, undotted i, and long f. 
Titles, where wanting, are added for convenience. The pronun- 
ciation is adapted to a slightly amended text, as the manuscripts 
are often very faulty, but the different provincial characters are 
not disturbed. The whole writing and versification is very rude 
and uncouth. 

MS. Cotton Cleop. B. 201 v°. 
Rel. Ant. 1, 22. 
Pater noster. 

CTre fadir )7at hart in heuene, 
halged be ])i name with giftis seuene 

samin cume J^i kingdom, 

]?i wille in her))ei als in heuene be don, 

vre bred j^at lastes ai 

gyue it bus J^is hilke dai, 

and vre misdedis })u forgyue hus, 

als we forgyue )>aim J^at misdon hus, 

and leod us in tol na fandinge, 

hot frels us fra alle iuele ]?ing. 


^ The printed text of the Eeliquice 
Antiquce was first read by me with the 
MSS., and the proofs of these pages 

Conjectured Pronunciation, 

Paa-ter nos'ter. 
UuTe faa-der dhat art in Hev-ene, 
Hal-ghed bee dhi naa-me with gift'is 

Saa'min kuu-me dhi kiq-doom-. 
Dhi wil in erth, als in hevne be don. 
Uu*re breed, dhat last-es ai, 
Giiv it us, dhis ilk-e dai, 
And uu're misdeed-is dhuu forgiive us 
Als wee forgiiv-e dhaim dhat miis-doon- 
And leed us in til naa fan-diq-e, [us. 
But freels us fra al iiv-le thiqe. 


were again compared with the originals 
by Mr. Brock. 

§ 1, No. 3. 




JTeil Marie, ful of grace, 
be lauird ]?ich j^e in heuirilk place, 
blisced be ]>u mang alle wimmem, 
and blisced be j^e blosme of ]>i wambe. 



JEi true in God, fader hal-micht- 
tende,)'at makedeheuen and herdej^e, 
and in Ibesuc Krist, is ane lepi sone, 
hure lauerd, bat was bigotin of ]>e 
hali gast, and born of j^e mainden 
Marie, pinid under Punce Pilate, 
festened to ]7e rode, ded and duluun, 
licbt in til belle, ]>e j^ride dai up 
ras fra dede to line, stegb in til 
heuenne, sitis on is fadir ricbt 
hand, fadir al-waldand, he j^en 
sal cume to deme Ipe quike an 
jje dede. Hy troue hy J^eli 
gast, and hely kirke, pe samninge 
of halghes, forgifnos of sinnes, \'p- 
risigen of fleyes, and life wi]7-hutin 
hend. Amera. 

A a* V e 

Hail, Marii-e, ful of graa'se, [plaa-se 
Dhe laa-vird thi^-h dhe in evrilk 
Blis-ed be dhuu maq al*e wim-en- 
And blis'ed be dhe blosm- of dhi 
wamb. Aa'men-. 

Kr e e- do 

li troou-e in God, faa-der al'mi^ht*- 
end-e, dhat maa'kede Heven and erth-e, and 
in Dzhee'sus Krist, His aa*neleep*i soo'ne, 
uu-re laverd, dhat was bigot-en of dhe 
Haa-li Gaast, and born of dhe Mai -den 
Marii-e, pii-ned un-der Puns-e Pilaat'e, 
fest'ened to dhe roo-de, ded and dulven, 
li^ht in til nel'e, dhe thrid-e dai up- 
raas" fra deed-e to lii've, stee^h in til 
Hcvene, sit'es on His faa'der ri^ht 
Hand, faa-der al-wald-and-, nee dhen 
shal kuu-me to dee-me dhe kmk-e and 
dhe deed-e. li trou-e [in] dhe Haa-li 
Gaast, and haa-li Kirk-e, dhe sam-niq-e 
of nal-ghes, forgif-nes of sin-es, up'- 
rii-si^/hen of flaish, and lii-ve withuuten 
end-e Aa-men. 

Sari. MS. 3724, fo. 44. Hel. Ant. 1, 57. Camden's Remaines, p. 24. Lyttelton 

History, 4, 130. 

Pater noster in Anglico Paa-ter nos'ter 

Vre fader in heuene riche, 
])i name be haliid euer iliche 
J7U bringe vs to Y\ michil blisce, 
Y\ wille to wirche \xi vs wisse, 
Als hit is in heuene i-do 
Euer in eorjie ben hit al so, 
]>at holi bred J^at leste}? ay 
hu send hit ous ])is ilke day, 
Forgiue ous alle j^at we haul)' don, 
Als we forgiuet uch oj^ir man 
Ne lete vs falle in no fondinge, 
Ak scilde vs fro j^e foule j^inge. 



I bileue in God fadir almichty, 
sshipper of heuene and of eor)?e, and 
in Ihesus Crist, his onlepi sone, 
vre louerd, J^at is iuange ^urch )>e 
holy gost, bore of Marie Mayden, 
]7olede pine vnder Pounce Pilat, 
picht on rode tre, ded and yburiid, 
licht in to belle, )?e jjridde day fram 
deth aros, steich in to heuene, sit on 
his fadir ricbt honde, God almichti, 
)7enne is cominde to deme j^e quikke 
and Jje dede. I bileue in J^e holy 
gost, al holy chirche, mone of 
alle halwen, forgiuenis of sinne, 
fleiss vprising, lyf wi]?uten ende. 


Uu're faa-der in hev-ne riitsh-e, 
Dhi naame be nal-jcd ever iliitsh-e 
Dhuu briq us too dhe mitsh-el blis-e, 
Dhi wil-e to wirtsh-c dhuu us wis-e, 
Als nit is in hev-n- idoo- 
Ever in erth-e ben it al-soo", 
Dhat Hoo'li bred dhat lesteth ai 
Dhuu send nit us dhis ilk-e dai, 
Forgiiv us al dhat wee navth doon, 
Als wee forgiv-eth eech ooth-er man, 
Nee leet us fal in noo fon-diq-e, 
Ak shild us froo dhe fuu-le thiq-e. 


li bileev in God, faa-der al-miA:ht'i, 
ship-er of nev-ene and of erth-e, and 
in Dzhee-sus Krist, His oon-leep-i soo-ne, 
uu-re loverd, dhat is ifaq-e thurkh the 
Hoo-li Goost, boo-ren of Mariie mai-den. 
thoo-lede pii-ne un-der Pun-se Pilaat", 
pi^ht on the roo-de tree, deed and iber-ied, 
liArht into nel-e, dhe thrid-e dai from 
deeth aroos*, staiA:h into nev-ene, sit on 
His faa-dir ri^ht Hond-e, God al-mi^ht-i, 
dhen-e is kuum-end-e to deem-e dhe kmk-e 
and dhe deed-e. li bileeve in dhe Hoo-lt 
Goost, al Hoo-li tshirtsh-e, moon-e of 
aPe Hal'wen, forgivnes of sin-e, flaish 
uprii-siq-, liif withuu-ten end-e. 




Chap. V. 

Arund. MS. 292, fol 3. Rel.Ant. 1, 234. 

Pater noster. 
Fader ure "Satt art in heuene blisse 
'Sin hcge name itt wur'Se bliscedd, 
Cumen itt mote 'Si king dom, 
"Sin hali wil it be al don 
In heuene and in er'Se all so, 
So itt sail ben fill wel ic tro ; 
?if us alle one 'Sis dai 
Vre bred of icbe dai 
And forgiue us ure sinne 
Als we don ure wiSerwinnes ; 
Leet us noct in fondinge falle, 
Ooc fro iuel 'Su sild us alle. 


AuE Maria 
Marie ful off grace, weel de be, 
Godd of heuene be wi'5 'Se, 
Oure alle wimmen bliscedd tu be. 
So be ^e bern datt is boren of ^e. 

Credo in Deum 
I leue in Godd al-micten fader, 
"Satt heuene and er^e made to gar ; 
And in Ihesu Crist his leue sun, 
Vre onelic louerd, ik him mune, 
"Satt of de holigost bikennedd was. 
Of Marie 'Se maiden boren he was, 
Pinedd under Ponce Pilate, 
On rode nailedd for mannes sake ; 
•Sar 'Solede he deadd widuten wold. 
And biriedd was in de roche cold, 
Dun til belle licten he gan, 
"Se "Sridde dai off deadd atkam, 
To heuene he steg in ure manliche, 
"Sar sitte'S he in hijs faderes riche, 

domes dai sal he cumen agen, 
To demen dede and lines men : 

1 leue on '5e hali gast, 
al holi chirche stedefast 
Men off alle holi kinne, 

And forgiuenesse of mannes sinne, 
Vprisinge of alle men. 
And eche lif I leue. Amen. 

Camden's Remnines p. 24. 
Paa'ter nos'ter. 
Faa'der uu're dhat art in nevne blis'e 
Dhiin Hekh*e naam it wurdh-e blis-ed, 
Kuu-men it moote dhii kiq-doom* 
Dhiin naadi wil it bee al doon 
In Heven and in erth al soo, 
So it shal been ful wel ik troo, 
Gif us aPe on ' dhis dai 
Uu're bred of iitsh-e day 
And forgiv us uu-re sin*e 
Als wee doon uure wiidh-erwin'es ; 
Leet us nokht in fon'diq*e fal'e, 
Ook fi'o ii'vl dhuu shild us aPe. 


Aa* ve 
Marii-e ful of graa*se, wel de^ be, 
God of Hevne bee with dhee, 
Ovr- aPe wim-en blist tu^ bee. 
So bee dhe bern dat-s^ born of dhe. 

Kr e e* do 
li leev in God aPmiA-ht-en Faa'der, 
Dhat Hevn-and erthe maad togaa-der; 
And in Dzhee'sus Krist, His lee've suune, 
Uur oo'neliik loverd, ik nim muu-ne, 
Dhat of dhe Hoo-li Goost biken-ed was, 
Of Marii-e dhe mai'den boorn ne was, 
Pii'ned un-der Puns-e Pilaate, 
On roo'de naiPed for man-es saa'ke. 
Dhar dhoold -e death withuu-ten woold, 
And ber'ied was in dhe rotsh'e koold, 
Dunn til nePe li/:ht-en ne gaan, 
Dhe thrid-e dai of death atkaam*, 
To Hevn -e steeyh in uur man lii'tshe, 
Dhar sit'eth -e in -is faa-dres rii'tshe, 
doo-mes dai shal -e kuu'men agen* 
To dee-men deed and lii-ves men. 
li leeve on dhe Haadi Gaast, 
Al-Hoo-Ii tshirtsh'e stee'defast, 
Men of aPe hoodi kin-e, 
And forgivnes' of man-es sin'e, 
Up"riis-iq* of al"e men. 
And ee-tshe liif ii leev. Aa*men*. 

The short Peoverbial Yeeses, voP ii, p. 14, are taken from the 
margin of the Cott. MS. Cleop. C. vi, fo. 21, where they are in a 
different hand from the text and are probably much later, though, 
as Mr. Wright observes, *'in a hand of the thirteenth century." 
They contain some peculiarities as : feise midoutin losing, for : J^ese 

1 This line is probably corrupt. The 
hiatus (aPe on), is unlikely, but to 
read : (Gif us aloo*ne . . . dhis dai), 
would be deficient unless we inserted 
(nun) or some such word, after (aloo'ne), 
meaning : give us alone [now] this 
day. The rhyme is, however, so rough, 
that criticism is out of the question. 

2 (De) for (dhe) after (wel); (tu) 
for (dhu) after (blist) which must be 
taken as a monosyllable, this change 
of (dhu) into (tu) shewing that the 
preceding letter was voiceless, that is 
(t) not (d), as j^u would have otherwise 
been (du), compare the first case, and 
also (dat) for (dhat) after (bern). 


wi]7uten lesing = these without lying. This form ]>eise is not named 
by Stratmann, and is perhaps an individuality. The ow in : midoutin 
stroutende, belong to the transition period, shewn distinctly by : 
** that tu, and tou,^^ both of which = ]fu, in two following lines. 
The form ulu, printed jelu, for -^elu = yellow, is peculiar, as 
shewing the complete passage of 5 into i. 

In vol. i, p. 89, there is a Hymjst to the Viegin", and another on 
p. 102,^ preceded by a curious parabolal poem, beginning: '' Somer 
is comen and winter^ gon," not entirely legible, all taken from 
MS. Egerton 613, fo. 1 and 2. The first and last are in the same 
hand, the second in a different hand, but they all belong to the 
transition period ; thus on p. 89 we have : thou, our, flour, ous 
(twice), foule ; but also : hut = out, thu (3 times) ; also : put = 
pit, shewing the (y) or (i, c) sound of u. The last has : foules == 
fowls, witoute = without, ous = us, but generally keeps the u 
pure. And the second prayer p. 102, while it has : thu (16 times), 
flur, withuten, orcisun, tunge, has also : out, foul ; and : sunne 
(3 times) = sin. It is curious to note also : ic chabbe, and ich 
chabbe, for ich habbe, implying probably the running on of the 
words thus : (i,tshab*c). The orthography : flehs, for : flesh, is 
perhaps to be compared with : ihc, for : icli, in the preceding line. 

The other poems in the Reliquio} Antiqiice, belonging either to 
the transition or later periods, do not call for any further remark. 

The first seven pieces in the Early Eiiglish Poems taken from 
Harl. MS. 913, are all assigned to a date prior to 1300, but like 
the fifteen pieces which follow from Harl. MS. 2277 and ascribed 
to 1305-10, they belong to the transition period with respect to 
ou and u. 

In the Saeivtutt pp. 1-7, the transition period is marked by : ous 1 
(the figures refer to the stanzas), nou 2a, mou]? 4, aboute 4, fou 5, 
wifoute 7, etc. ; against : urc 1, us 3, schuldres 5, luse = louse 5, 
wijjoute prute = proud 6 (the adjective always end in t ; prude 10, 
pride 12, is the substantive in which w = «',) acuntis 24, lude = 
loudhj 31, jur 41, etc. The u for i is common, as munde kunde = 
mind kind 26, ihuddid 11. The palatalised guttural usually sinks 
into i, as : seij? 3, mei 8, dai 18, ei hei = eye high 22, etc. ; but j 
sometimes remains, as : hei^ 53, 56, ne^bor 9 ; ])ei} = though 27. 
We find also : fleisse meisse = flesh mass 6 (see infra p. 473, n. 4), 
hir hirist = herr, hearest 33, file = vile 3, drit = dirt 7, dritte = 
dirt 10, ihc 13, mov = mow 14, nov = now 31, verj^ing = farthing 
24, wl = will 31, angles = angels 33, woni = to dwell 51, and 
these infinitives in i, usually accented, occur as will be presently 
seen, in other parts of the same MS. There is an assonance : sprede 
wrekke 30, and : virst best 57 may probably be : frist = thirst 
best, a rhyme of i, e, but the rhymes in general are not remarkable. 
The final e seems simply disregarded in rhyme and metre, but the 
metre is so hummocky that it is difiicult to make anything of it. 

1 Both are printed in Goldbeck and ^ The is here inserted in the printed 

Matzner's Altenghsche Sprachproben, text of the Eel. Ant. is not in the MS. 
p. 53. 


Take for example the last stanza, p. 7, wliich may perhaps be read 

as marked : 

Alle ])at bej7 icommin here (Al dhat beeth ikum-en neer 

fort to hire j^is sarmun For to nii-re dhis sarmuun-, 

loke yat je nab no were Look-e dhat je n-ab no weer-e. 

for seue jer je habbi]? to pardoun. For sev jeer je nabth parduun :) 

The whole MS. seems marked by provincialisms, which it is ex- 
tremely difficult to understand. The first stanza of the xv. Signa 
ANTE JUDICIUM, p. 7, is in the same style, and was probably due to 
the same author : 

pe grace of ihfsu fulle of mijte (Dhe graas of Dzhee-su ful of mikht 

])roj prier of ure swete leuedi Thrukh prii'er of uur sweet levdii* 

mote amang vs nuj^e alijte Moot amaq- us nuudh alikht 

And euer vs jem and saui. And ever us jeem and saavii-) 

Such attempts, however, to give pronunciation, riiust be viewed 
with indulgence, they are necessarily very hazardous. In this 
piece : ysaie profecie 9, must have the vowels in at divided, y-sa-i-e. 
The final e in mercie 25 is idle, added on to rhyme with crie in the 
same stanza, where it was probably not pronounced, as we have : 
of ih^su crist merci to cri 80, and 

])e. xii. dai j^e fure. elemens sul cri 

al in one heij steuene 

merci ihesu fij mari 

as ];ou ert god and king of heuene, 177 

which gives us another example of Mari, see supra, p. 441, and 
similarly: to cri, merci 137. Eemarkable forms: dotus angus = 
doubtful avguish ] 13, probably = (duutus' aqgus*) with a JSTorman 
u = (u), fisses =Jishes 121, euch uerisse watir = each fresh water 
125, skeis = skies 133, where I suspect an accidental transposition 
of ei for te, as the form is otherwise incomprehensible, fentis = 
fiends 161, fure = four 169, 177, wolny nulni = wullen-hi ne- 
wullen-hi, = will they nHll they 173, maugrei = maugre 173, pro- 
bably a JSTorman form. 

The Fall and Passion, p. 12, has the rhyme: frute dute=" 
fruit doubt 23 (line) which is decidedly favourable to the English 
pronunciation of ]^orman u at that time as (uu) see p. 424, note 3. 
Kemarkable forms : maistre = mastery 21, maistri = mystery 50, 
sso = she 52, ^o = she 79, flees, = flesh 49, as he is manhed siwed 
97, hou hi lord ssold siu fe 105. The following infinitives in -i 
occur : suffri = to suffer 66, honuri = to honour 72, biri = to bury 
74, 76 ; and : sauid isinid 43, being accented on the last syllable 
imply the same form. The same accent occurs in the rhyme : 
ipinsed suff'red 89, siwed suffrid 97. The rhyme : alowe two 79, 
seems to be an error. 

The Ten Commandments, p. 15, has also: honuri worfi = to 
honour, to worship 17, and the assonance : iwisse limmes 5. The 
Peagment on the Seven Sins, p. 17, has also : clansi = cleanse im- 
perative st. 5, herrid = AornV?, st. 10, nemeni = ^o name st. 10, 
woni = to dwell v. 9, prute shrute = proud shroud v. 10, fleis = 
flesh V. 12, Jer is mani man bi peijte (= bepeachedy deceived ?)j so ];e 


fend Mm hamf itei^te {= taught '^) 22, susteni = to sustain 58. 
Chkist on the Cross p. 20, has : bewoncle wnde = wounded wound 
V. 3, fote blode 11, anguis 14, gredind deiend 25, Strang hond 26. 
The Rhyme beginning Feagment p. 21, is only remarkable for 
making in me answer to inne, but as the trick of beginning a line 
with the last word of the preceding line is not carried out con- 
sistently, this assonance may have no special meaning. The whole 
examination does not lead to much. The orthography is so singular 
and so irregular, we might almost say so ignorant, and the dialect 
so peculiar, that it is of very little assistance. No general result 
could be deduced. The rhymes are not certain enough to be of 
much value, and are generally the veriest doggrel conceivable, 
while the metre is nowhere. In the parts from Harl. MS. 2277, 
we may notice the false rhymes : poynte queynte p. 66, v. 5, 
(unless mOiQQdi poynte is to be Normanized into peynte), britaigne fawe 
p. 68, V. 85, against : britaigne fayne p. 69, v. 133, and the asso- 
nance : makede glade p. 108, v. 35. The form sede for seide is 
found in : rede sede p. 66, v. 28 ; p. 68, v. 99, sede mede p. 72, 
V. 56, dede sede p. 74, v. 48. See infra, p. 484. But seide also 
occurs, p. 72, v. 58, v. 60, etc., being the regular form. 

In an extract from Cott. MS. Vesp. D. IX., (which being of the 
XV th century, does not properly belong to this place), Why I 
can't be a nifn, p. 138, we find : wept few accept ihesu trew ob- 
servaunce new variaunce p. 139, v. 40, but ihesu may not have been 
intended to rhyme y^ith. few trew new, because we find a line ending 
with this name thrown in without a rhyme on the next page 140, 
V. 88, kyn neccssite wyn me omnipotent Ihesu present ys thys, etc. 
In p. 140, V. 100, we find: 

To the for comfort I make my sute 
To have that ioy that lastythe ay. 
For her loue that bare that frute 
Swete ihesu miserere mei. 

giving the rhyme : ay mei, the last word being Latin : This may 
be compared with : Sinay day, in Chaucer, supra p. 264, and Dr. 
Gills (eei)p. 114. 

In the Political Songs Mr. Wright has collected a number of short 
poems in Latin, JSTorman French, and English, referring to the xin th 
or beginning of the xiv th century. Unfortunately most of the 
English songs, as : the Song against the King of Almaigne p. 69, 
Song of the Husbandman p. 149, Song against the Pride of the 
Ladies p. 153, Satyre on the Consistory Courts p. 155, Song of the 
Flemish Insurrection p. 187, Execution of Sir Simon Eraser p. 212, 
Song against the Retinues of Great People p. 237, Elegy on the 
Death of Edward I. p. 246, are from Harl. MS. 2253, which has 
adopted the full xivth century orthography, so that they are of 
little use here. The principal points are the assonances : lonke 
songe wlonke thonke p. 156, and longe londe p. 193. There are 
numerous instances of w = (i, e), as : hude prude p. 150, sturne 
hurne p. 150, wunne sunne p. 153, prude shi^ude hude p. 153, 
prude drede p. 190. The apparent rhyme : ded sayde p. 246, 




Chap. V. 

is probably no rhyme at all, but the nature of the stanza is broken 
and the first and third lines do not rhyme, which is precisely what 
we find in the next stanza but one, p. 247, where othermse : 
sunne Edward, would form a rhyme ! Still, as we have just seen, 
the form sede also occurs, and may here be meant (p. 447). Ded 
gret redde p. 248 must be regarded throughout as an assonance. 
In : chivalrie deye hey go crie p. 249, the second and third words 
should have been written : dye hye, as often in Chaucer. 

Of all the Political Songs the only two which exhibit almost pre- 
cisely the orthography of the xiii th century, are those from the 
Harl, MS. 913, viz. The Song of the Times, p. 195, and the Song of 
Nego, p. 210. The last raises no new points, and may be passed 
over. The first exhibits ss for sch in : ssold p. 197, also written 
schold in the same page, ssal pp. 201, 203, 204, ssul pp. 202, 205, 
precisely as in the Ayenlite, supra p. 409. There are some little 
slips as: feloni = felonie p. 197, line 13, amy lie, ami mei both 
on p. 200, where 7nei is an error for me. The first will not rhyme 
unless we read : U\ which is unusual, but the fijial ^'s are lax in 
this song. The use of hoi = boy, in : tel me, boi, what hast 
ido ? p. 199, 1. 5 is noteworthy. The curious word i-pilt, in the 
Prisoner's Prayer, v. 25, (supra p. 429, note 1), is well illustrated 
by the passages 

And so men didde that sell asse, 

That trepasid nojt, no did not gilte, 
"With ham bothe iwreiid was, 

And in the ditement was ipilt. p. 198. 
Godis grame most hi have 

That in the curte the so pilt ! 

When hit is so, ich vouchsave, 
Ic forgive the this gilte. p. 

Ic am iwreiid, Sire, to the, 
For that ilk gilt ; 

Sire, ichul sker me, 
I ne jef ham dint no pilt. p, 



The Auchinleck ^ MS. in the Advocates' Library, Edinburgh, has 
not thoroughly adopted the xiv th century orthography,^ and as it 
belongs to the very beginning of the xivth century^ has a claim to 

1 "In the year 1504, the barony or 
manour of Auchinleck (pronounced 
Afflecl') in Ayrshire, which belonged to 
a family of the same name with the 
lands, having fallen to the crown by 
forfeiture, James the Fourth, King of 
Scotland, granted it to Thomas Bos- 
well." — Boswell's Life of Johnson, 
anno 1776. "The pronunciation of 
Affleck for Auchinleck, was formerly 
common, but is fast disappearing, and 
is now confined, I should say, to the 
lower classes of the parish and neigh- 
bourhood." Private letter from Mr. 
Halkett, Librarian of the Advocates' 
Library, Edinburgh, 18 Jan. 1869. 

2 Nu, hu, occur occasionally, but 
rarely. Nu occurs once in the piece 
immediately cited, nu and hu several 
times in the second piece, which, though 
last in the MS., is said to be in an 
older hand. I have not noticed any 

such forms in Sir Tristrem, the 37th 

^ An "Account of the Auchinleck 
MS. Advocates' Library (W. 4, 1,) and 
a catalogue of its contents," forms the 
fourth appendix to the introduction to 
Sir Walter Scott's edition of Sir Tris- 
trem, to which a facsimile of the first 
two stanzas of that poem are prefixed 
It is a quarto of 334 leaves, containing 
44 pieces of poetry, on parchment, " in 
a distinct and beautiful hand, which 
the most able antiquaries are inclined 
to refer to the earlier part of the xiv th 
century. The pages are divided into 
two columns, unless where the verses, 
being Alexandrine, occupy the whole 
breadth of the quarto. In two or three 
instances there occurs a variation in 
the handwriting ; but as the poems re- 
gularly follow each other, there is no 
reason to believe that such alterations 

§ 1, No. 3. 



be considered here. There are two extracts from it, On the King's 
Breaking his Confirmation of Magna Charta, p. 253 (MS. l^o. 21), 
and the Evil Times of Edward II. p. 323 (MS. JN'o. 44). The 
second only ofi'ers the curious orthography : muis huis, p. 326, 
for : mous hous, and the assonance : hundred wonder p. 344. 
Eut the first is very singular. The second, third, fourth, fifth, 
and sixth stanzas, containing the sayings of the "iiij. wise men" 
have a peculiar arrangement of rhymes, differing from the rest of 
the poem, which may be symbolised thus, like letters shewing 
rhymes : a a h cc h ddd eee. The last five stanzas stand thus : 
a a h c c h. I^one of these lines present any difficulty or novelty. 
The following is the first stanza, which Wright prints in divided 
lines, but which in the MS. itself runs across the page, although the 
pages of the MS. are usually divided into two columns, indicating, 
apparently, that the transcriber considered the final rhymes only as 
pointing out the divisions. 

Len puet fere et dcfere ceo fait il trop souewt 

It nis nouj^er wel ne faire yerfore engelond is shent 

Noftre prmce de engletere par le confail do sa ge«t 

At weHminRer after l>e feire maden a gret parleraewt 

La chartre fet de eyre ieo lewteink et hien le crey 

It was holde to neih ])e fire and is molten al awey 

Ore ne say mes que dire tout i va atrz'polay, 

hundred, chapitle. court an shire al hit gop a deuel way ^ 

des plusages dc latere ore cfcotez vn sarmouw 

Of iiij. wise mew ])at j^er were whiengelonde is brouht adoun ^ 

indicate an earlier or later date than 
may be reasonably ascribed to the rest 
of the work ; although the satire 
against Simonie, No. 44, seems rather 
in an older hand than the others, and 
may be an exception to the general 
rule. The MS. was presented to the 
Faculty of Advocates, in 1744, by 
Alexander Boswell, of Auchinleck, a 
Lord of Session, by the title of Lord 
Auchinleck, and father of the late 
James Boswell, Esq., the biographer of 
Dr. Johnson. Of its former history 
nothing is known. Many circum- 
stances lead us to conclude that the 
MS, has been written in an Anglo- 
Norman Convent. That it has been 
compiled in England there can be but 
Httle doubt. Every poem which has a 
particular local reference, concerns 
South Britain alone .... On the other 
hand, not a word is to be found in the 
collection relating particularly to Scot- 
tish affairs." 

1 Compare "And lete me slepe, a 
twenty devel way !" — Cant. Tales 3713. 

2 The passage as we learn by Mr. 
"Wright's note on p. 385, was trans- 
ferred to his pages from : "an in- 
teresting little volume of early poetry, 

edited and printed privately by David 
Laing, Esq., and "W. B. D. Turnbull, 
Esq., under the title of ' Owain Miles, 
and other Incdited Fragments of An- 
cient English Poetry. 8vo. Edinburgh, 
1837.' " The present copy follows a 
careful transcript obligingly made for 
me by Mr. Halkett, the Librarian of the 
Advocates' Library, Edinburgh, who 
says : "In Owain Miles the editors 
have divided each line into two ; in my 
transcript you have them as they were 
originally written. There are no points 
except a dot after the word hundred, 
and another after the word chapitle ; I 
am not sure whether they have been 
put there by design or by accident." 
On examining the orthography of the 
two pieces in this MS. given by 
Wright, and of Sir Tristrem as edited 
by Scott, we find it very irregular with 
respect to final e, in which it agrees 
with the MS. of Hampole (p. 410). 
Similarly, in the poems of the "deeff, 
sick, blynd," monk John Audelay of 
Haughmond, four miles from Shrews- 
bury, written 1426, necessarily from 
dictation and of course unrevised by 
the author (edited from MS. Bodl. 546, 
for the Percy Society, by J. 0. Halli- 




Chap. V. 

Now if wc adopted Mr. Wright's arrangement in half lines 
we should be led to suppose that the rhymes were intended 
to be arranged thus : ab ah a I ah cd cd cd cd efefj 
and thus make : defere faire Engletere feire, rhyme together. But 
the first and third words probably ended in (-ee*re) and our pre- 
vious investigations lead us to consider that the second and fourth 
ended in (ai're).^ We have not hitherto found a single instance in 
any good xiv th century MS. of e rhyming with ay or ey? The few 

well, 1844), the final e has appa- 
rently no phonetic meaning at all. The 
whole character of the spelling of Sir 
Tristrem (MS. No. 37) is northern. 
In the present short extract we have 
both Engelond and Engelonde in the 
nominative ; in the second line faire 
should be fair (p. 383), and then of 
course feire must be feir^ if it is in- 
tended to rhyme, maden a gret parle- 
ment, seems an error for, made a grete 
parlement ; Wright reads made a gret. 

* A Somersetshire farming man once 
asked me if I had seen the (shep) on 
the (fair), which sounded remarkably 
like a ship on fire., but merely meant 
the sheep in the fair from which I was 
walking. This is therefore an existent 
(fair) pronunciation of the Norman 

2 This rhyming of ey with ^, must 
be distinguished from the double forms 
ey^ e, in certain words which occurs in 
a few instances, see supra p. 265, and 
compare the double forms ey., i, pp. 
284-286. The apparent rhyme : ded, 
sayde, p. 448, we have seen may not 
have been meant for a rhyme at all. 
Since the text was in type my atten- 
tion has been directed to some apparent 
rhymes of ay, e in the poems of Aude- 
lay referred to in the last note but one. 
It will be advisable to consider these 
rhymes in this place. We must re- 
member that the poet was both blind 
and deaf, and had an illiterate scribe. 
These three points are well proved by 
the notice (p. vi., Halliw ell's edition, 
to the pages of which all references 
will be made) : " iste liber fuit compo- 
situs per Johannem Awdelay, capella- 
num, qui fuit sectis et surdus, in sua 
visitacione, ad honorem Domini nostri 
Jhesu Christi, et ad exemplum aliorum, 
in monasterio de Hagraon, anno Domini 
millesimo cccc^^^ vicessimo cujus 
anime propicietur Deus." The secus for 
cecus., or as we now write ccecus, shews 
the trustworthiness of the scribe. The 
English part is full of the grossest or- 

thographical eccentricities and incon- 
sistencies, and was probably written by 
an ignorant brother, whose labours the 
author was unable to revise either by 
eye or ear. Under these circumstances 
we should rather be surprised at the 
regularity of the rhymes than at the 
occasional utter forgetfulness of rhyme, 
as : law withdrawe dais p. 22, (but 
perhaps dawes should be read, see supra 
p. 371, Ex. b.), leudmen corexeon 
relegyon p. 24, Christ charyte p. 26, 
to therfro more p. 40, worlde Lorde re- 
warde p. 40, reprevyd dispilid p. 60 (both 
accented on the penultim), Lorde worlde 
p. 60, Judas cos = kiss p. 60, Lord 
soiFyrd p. 61, thorst last p. 64, opus 
masse p. 73, on-morwe undorne p. 75, 
dimes masse p. 76, dynt stout p. 78, 
masse worse p. 79, prayere honoure 
p. 79. It is evident from these ex- 
amples that we must not press Audelay's 
doggrel rhymes too closely, and cer- 
tainly not draw any inference from a 
few isolated examples. There can be 
no doubt, however, that he did not dis- 
tinguish short i and short e, and there 
seems little doubt that he confused long 
i and long e also. Every page offers 
examples of the first, and the rhyme in 
-e, -i, -y, -ye is the commonest he has. 
The words, die, high, eye, were to 
him dye, hye, ye the last was even 
written -e, (p. x), and the two former 
constantly rhyme -e. Mr. Halliwell 
says (p. xi) that in Shropshire " i is 
still turned into e, which may be re- 
garded one of Audelay's dialectical 
peculiarities, especially in the prefixes 
to the verbs." Another peculiarity, of 
the scribe at any rate, is to consider oi 
and i as identical, at least in some 
words. We have already cited dispilid 
= despoiled, p. 60, and we have dystry 
p. 20 but dystroy p. 33, foyre =fire, p. 
48, rhyming to were. Another singular 
rhyme, if any weight is to be attached 
to it is: hyng drynke p. 61, see supra 
p. 192. The word cros has various 
rhymes : was, losse p. 61, choys p. 8, 

§ 1, No 3. 



earlier cases which appear to exist in Havelok, etc., will hereafter be 
shewn to have probably arisen from errors (p. 473). Could we then 

were it evidently stands for croyse 
which is used p. 64 to rhyme with 
voyse. This preliminary examination 
will enable us to appreciate the exam- 
ples of ay which apparently rhyme 
with e. In the first place, although 
-e, is the commonest rhyme sound 
throughout the poems, and -ay is also 
frequent, the instances in which -ay 
rhymes with -e are very rare. The 
following are all that I have noted 
throughout the extracts edited by Mr. 
Halliwell. In the poem on Henry VI, 
p. viii, there are 16 lines which should 
rhyme in ay, but in one case the word 
is cuntre, the rhymes being : veray day 
play away fray day way day aray day 
cuntre Fryday may betray pray Awd- 
lay. The rhymester was evidently 
hard up, or he would not have used 
day five times, and if his ay had really 
rhymed to his common e, he would 
certainly have introduced it many times. 
The single instance might be a case of 
carelessness, which the blind and deaf 
man failed to discover and correct. 
But country is one of those words which 
had a double orthography : cuntre 
cuntrey, corresponding to two forms 
in the Norman, which generated two 
pronunciations in {-i -e) in xvii th 
century (supra p. 125), and hence pro- 
bably had two sounds (-ee, -ai) in the 
XIV th century at least. To this list 
belong: country, valley, journey, livery, 
most probably. Hence the error may 
be merely scribal, cuntre for cuntrey. 
Cumpane, which at first sight seems to 
rhyme with say, p. 16, is apparently a 
simple mistake, and the line containing 
it, which is unnecessary to the sense, 
should be expunged. It occurs among 
a set of 78 stanzas of 13 lines each, 
having the complicated rhyme system 
ababbcbcdeeed. In this par- 
ticular piece the rhyming words are : 
spiryt say epocryte pay day compane 
clene say lene mynde by truly cumpany 
unkynde. That is, this one stanza has 
14 lines; and the line which is subver- 
sive of the whole rhyme system, is this 
very one which ends in compane. 
Degre be may p. 44 is also a mere 
error, it occurs in a stanza of the 
last kind, corresponding to the e e e 
portion, and on the same page, in the 
next preceding stanza, in the same por- 
tion, we find : jeve know laue, which no 

one would hesitate to consider a false 
rhyme. To the same category I relegate 
the example in the same place of the 
next stanza : sayne eyne sene p. 45, 
where sene = seen is the infinitive mood 
oi see, y-seyne bene p. 68 =i-seen been, 
are past partici pies, and the spelling 
of the first word is erroneous, but we 
have a similar form in Chaucer, supra 
p. 265. Bred betrayed p. 70, I class 
with : wayt algat p. 47, as mere helpless 
rhymes ; if the one could prove that 
ay = (ee), the other would prove 
ay = (aa), for the rhyme : face alas p. 
60, would establish longa = (aa). In 
cownsele asayle batayle p. viii, the first 
word should have its usual form coun- 
seyl. In erne = modern aim p. 12, 37 
and often, the e is correct, the modem 
spelling is wrong, the origin being Fr. 
esmer = ajstimare. The above are ab- 
solutely all the cases observed, and the 
impression produced on myself by the 
examination of these rh)Tnes, is, that 
Audelay pronounced ay, e, differently, 
and that the conclusions deduced from 
other sources apply to these cases also, 
viz : ay = (ai), e = (e). Nevertheless 
there are at least two MSS. and there 
may be more, which certainly confuse 
cy, ay with e, both in spelling and 
sound.. The most striking of these in 
the Lincoln's Inn. MS. 150, from which 
Weber has printed the greater part of 
King Alisaunder (in vol. 1 of Metrical 
Romances of the xiiith, xivth and 
XV th centuries, published from Ancient 
Manuscripts, with an introduction, 
notes, and a glossary, by Henry Weber, 
Esq., Edinburgh, 1810, 3 vols., 8vo.), and 
which must be carefully distinguished 
from the Bodl. MS. Laud, 1. 74, from 
which he has taken v. 4772 — 5989 of 
the same romance. This poem is 
supposed to have been written before 
1300, and both the MSS. are attri- 
buted by Weber to the xivth century, 
but Mr. Furnivall and Mr. Skeat 
date the Lincoln's Inn MS. about 
1450. The Bodleian MS. has nothing 
strange, except : noye daye 5412, 
ryth nyth 4812 (but: nighth righth 
5076) which reminds one of Havelock's 
peculiar th, infra p. 477. and there are 
a few i, e rhymes, as : clere fire 5342, 
and some e, a, as : art cert 5802, but 
not frequent, and some assonances, as : 
blith wyf 5138. But on the whole it 



Chap. V. 

from this popular song conclude that all this is a mistake, and suppose 
that Chaucer, and Gower, and other writers, although frequently 
hard up for rhymes, never employed such an extremely convenient 
jingle which lay ready to hand ? The conclusion would he hazard- 
ous in the extreme, and is certainly unnecessary, for the apparent 

is tolerably regular, and admitting the 
correctness of: cuntrey 4898, 6008, 
charrey 5096, curreye 5118, tornay 
play journay noblay 5212, presents no 
other remarkable orthography. But 
the Lincoln's Inn MS. is very peculiar, 
and if we had to deduce pronunciation 
from its rhymes, we should be badly 
off indeed. Omitting the false rhymes, 
63, 305, 1515, 1708, 3173, etc., the 
assonances, the cases in which the first 
syllable of a dissyllable rhymes with a 
monosyllable as : bridel ride 953, walles 
al 1876, foughte doughte 2761, certis 
heort 6544, etc., the rhymes of a with 
e, and even «, as: wist cast 716, 
fynde thousand 2403, often spelled 
thousynde, sixe waxe 6038 ; of e with 
t, «, ; and confining ourselves to the 
combinations ei, at, oi, ui or ey, ay^ oy^ 
uy, we find ei written for e in : leynthe 
streynth 788, 7351, nobleys 1373, 
eynde 1573, 1912, cleir 2885, steil 
3211, speide neide 3441, yeilded 3791, 
heynde 4206, yeir 6963, which are 
conclusive as to confusion in the scribe's 
mind between these sounds. But we 
also find ai rhyming with a, e, i, oi ; 
ei with ai, e, i, uy; oi with ai, i ; ui 
with e, i, oi. These rhymes are so 
curious that many of them may be 
cited. AI, A; saide made 525, 7339, 
barbicans mayne 1591, amiraylis talis 
1780, Taran, mayn 3247. AI, E: 
camelis vitailes 854, hoiiere=debonnair 
faire 6732, saide lede 6942, saide 
maied = mede? 7327. AI, I: Akaye 
Arabye 3399, play dye = deye 3442, 
bywryghen sayn 4116, raineth schyn- 
eth 6450, high contray 7143. AI, 01 : 
y-said anoyed 273, 876, 1287, 1599, 
and often, play boy 1730, {boy is ab- 
solutely written bay 4376), taile spoile 
2133, faile Tysoile 2148, palfray boy 
3207, pays = peace noise 3373. EI, 
AI : chevynteyn mayn 3199, reyne 
mayne 7378. EI, E .• thede feide 95, 
dey8 = dais nobles 1039, ese deys=dais 
1153, kene eyghnen 1317, yeilde sheldis 
2067, &eye=^seen pudre 2179, corteys 
■pes=peace 2951, yeld field 2959, steil 
wel 3419, keip=^keep deep 3429 (but: 
kepe deop 3477), seide felawrede 6838, 
mesteir conseiler 7480. EI, I : nygh 

fle'ygh 119, kynde heynde 425, yilde 
feilde 2956, is deys = is dais 3966, 
eighte knyght 3884, 6042, contrey 
sygh 6440, wite disseyte 7704. EI, 
UY : reyn iib\iyn = abide 2991. 01, AI 
see AI, 01. 01, I : annoyed distryed 
129, syghe joye 6060, nigh anoye 6116, 
anoye dye 6568. UI, E : kuyn=kine 
slen 760, quarter wildftiyr 1902, pruyde 
wede 2093, there afuyre 7549. UI, I : 
Tyre wildefuyre 3031. The conclusion 
seems to be that the writer occasionally 
pronounced a, ai, e, ei, i, oi, ui in the 
same way = (ee). This must certainly 
indicate some great peculiarity of pro- 
nunciation, and it is sufficient to note 
its inconsistency with the results al- 
. ready obtained. No more can be said 
than that some xv century scribes in 
some part of the country, did perhaps 
so pronounce. But I cannot think 
that these rhymes justify our supposing 
an invariable pronunciation of ai, ei, oi, 
ui in this manner by any speaker. 
There is another MS. Advocates' Li- 
brary Jac. V. 7, 27, supposed to have 
been written in the xv th century, from 
which Weber has printed his Sir 
Amadas (Ibid. vol. 3), which exhibits 
great peculiarities, of which we need 
only notice : reyr = rear 7, \eyt = let 10, 
geyt=get 24, deytte =debt 37, feyr = 
fere 118, grey t = ^r^a^ 156, seyt = se^ 
218, deyd reyd = deed rede 236, speyke 
meyte = s/»m/c meet 284, etc., shewing a 
complete fusion o^ei, e. The other pieces 
prmted by Weber, and all the other 
old spelling which I have examined 
are free from such fusion. The above 
peculiarities are also absent in the 
second copy of Sir Amadas printed in : 
Ghost-thanks or the Grateful Unburied, 
a mythic tale in its oldest European 
form, Sir Amadace, a middle North 
English metrical Romance of the 
XIII th century, reprinted from two 
texts with an introduction by George 
Stephens, Cheapinghaven (i.e. Copen- 
hagen), 1860, which Mr. Payne has 
brought under my notice. With this 
explanation, therefore, I allow the text 
to stand unaltered, convinced that al- 
though a few words may have had both 
(ai, ee), and a few provincials may 

§ 1, No. 3. NORMAN FRENCH EI, AI. 453 

anomaly is easily explained. The writer began in Worman Frencli, 
meaning to mix up English with it, just as Norman Prench, 
English and Latin are intermixed in a haphazard manner in the 
Song of the Times, p. 251. In this way he wrote the two first 
lines, taking the arrangement in the MS., (which did not rhyme in 
the middle) ; but reverting to Norman French in his third line he 
threw off a middle rhyme to his first, and then for the sake of 
symmetry he made his fourth line have a middle rhyme to his 
second, thus producing, if we count the middle rhymes, the some- 
what singular arrangement : ah ch ah ch. Naturally enough 
in adding the next four lines he adopted the more obvious ar- 
rangement : ah ah ah ah, for the words : eyre fire dire shire, 
all rhyme ; ^ and the words : crey awey Tripolay wcy,^ also 
rhymed to English ears at least, as (-ai). A question, however, 
arises whether the Norman French : crey, Tripolay, ended in (ai) 
as well as the English : awey, wey. Of the latter we can at 
present feel little doubt, of the former there may be considerable 
cause for hesitation. In modem French ei, ai, are in most words 
called (ee), and the stanza we have been considering has been relied 
upon to establish that ai, ei in English had the sound of (ee), on the 
presumption that : defere, faire, Engleterre, feire, were all intended 
to rhyme in (ee're).^ If we take the arrangement of the lines in 
the MS. itself, there is no room at all for this assumption, because 
in fact we have only ten rude Alexandrines, rhyming thus : a a a a 
h h h h c c, Sit their ends, and occasionally, but not essentially, 
rhyming their middles. A s, however, the other view is strongly 
insisted on, it is advisable, without further reference to an isolated 
song which can really establish nothing, to enter upon an examina- 
tion of the probable value of ei, ai, in old Norman, a question so 
extensive and so beset with difficulties that it is impossible to 
discuss it fully.* 

The conclusions to which I have been led by an examination of 
all the rhymes in Wace's Roman de Eou,^ and several other Norman 

have used (ee) for ei (ei, ai) in some (s/rre, dn'-re). The rhyme was there- 
words at a very early period precisely fore (-ii're) or (-erre) in all, or the first 
as Hart did in the xvi th century (p. in the French and the second in the 
122), the great majority of educated English. 

men, and all speakers of the Court 2 Mr. Wright prints wai/, Mr. Hal- 
dialect said (ei) or (ai) where ei, ai kett transcribes wei/. 
were written, down to the middle or 3 Rapp has adopted the pronuncia- 
^""^.f ?'!r,''''i*^ century and believ- ^^^^ u^f for ai in old French, see 
mg that the hypothesis of an origmal i^f^^'/^og note 1. 
(ee) sound, lollowed by an (ai)pronun- a c^ 1.1. • ^ .«r> 
ciation in the xvi th century as dis- ' ^^^ ^^^ P^^^^us remarks, p. 438. 
tinctly laid down by Sir T. Smith, (p. * Le Roman de Rou et des Dues de 
121), which again became (ee) in the Normandie, par Robert Wace, poete 
XVII th, is untenable. normand du xii® siecle, publie pour la 
1 Fire ]ias a, dsLiiYe e ; shire Sigs. scire premiere fois, d'apres les manuscrits 
an essential e. The word shire is still de France et Angleterre, avec des notes 
pronounced (shiir) by many, supra p. pour servir a 1' intelligence dutexte, par 
275, note 3. Cyre, dire, were French Frederic Pluquet, Rouen 1827, 8vo, 
(siir^, diir^) anglicized, perhaps to 2 vols., 16547 verses. 


poems, are that ei, ai, when written were always meant to indicate 
the diphthongs (ci, ai) or the dissyllables (e,i) and (a,i), but that 
they were occasionally employed, perhaps by a scribal error, for 
simple e (e). It also appears tolerably certain that in a small series 
of words both (ai) and (e) were pronounced at a very early period, 
and that in other cases, by the same sort of habit which at the 
present day leads an Englishman to terminate his {ee, oo) in (i,u), 
thus {eei, oou), and which led him in the last century to palatalise 
k, g into (kj, gj) before (ae, oi), — habits which, it is important to 
observe, exist in full force at the present day in Icelandic, the living 
representative of the language spoken by the JSTorsemen before they 
acquired Normandy, and therefore probably indicating the tendency 
of the pronunciation these would adopt — the Normans introduced 
an unhistorical, but really pronounced (i) after e, a, in many words ; 
so that this introduced i was not an idle orthographical ornament, but 
implied an actual alteration of sound. Whether the sounds (ei, ai) 
were kept as distinct as they now are in modem French conseil, 
travail, it would be difficult to determine, but they were certainly 
confused in writing, and it is probable that to English ears, which 
seem to have long confused the sounds, they sounded the same as 
the ordinary English (ai).^ The existence of the sounds (ei, ai) in 
vieil, ail and such words, seems indeed to imply a prior (ei, ai) pro- 
nunciation, because, as we have every reason to suppose that the 
palatisation of the / in (Ij) and even (jh) or its entire absorption in 
(i), as (vjei, ai), is comparatively modern, and we know that / had 
the contrary tendency to labialisation after the same vowels e, a ir 
Erench, compare eux, aux, it seems probable that this palatisation 
was generated by a preceding (i) and did not conversely generate 
the (i). Supposing these conclusions were correct, an English^ian, 
at least, would rhyme : crey awey Tripolay wey, as we have 
supposed, in (-ai). The following is a brief statement of some of 
the grounds on which these conclusions rest. 

Both ai and ei occasionally represent divided vowels and not diph- 
thongs, in which case the Erench editors generally write ai, ei, but 
it is more convenient to use the ordinary signs ai, ei with Dr. 

1 Modern Englishmen readily hear ei, ai are written ei, <b and pro- 
all combinations which approach in nounced {eei^ aa?) with a distinct 
sound to their (ai), as (ai). Compare and lengthened primary, and an 
p. 123, note 4. Observe the common extremely abbreviated secondary ele- 
error (kuu-dail) for (ku doei) coup d'ceil. ment. Compare the effect of the similar 
See also the various Scotch sounds, p. sound {eei) of southern English long a 
290, which Englishmen usually find at Tenby, p. 272, note 3. Also observe 
the greatest difficulty in distinguishing. the actual change of long a into (ei) or 
When I was recently endeavouring to (sei), as (rsein-e dsei) for rainy day, 
make a literary English friend appro- arnong the children of the uneducated 
ciate the difference (ei, ai), I entirely classes in London, pointed out to me 
failed, and he heard both sounds as by school-teachers to whom it occa- 
(ai). The Dutch< ei, ij =(ei, ai) as I «ioned difficulty, see p. 294 and note 2. 
heard them (p. 295, note 1), are both The change of (ee) into (ei) and thence 
heard as (ai) by Englishmen, and as (ai) is therefore not merely a priori 
(ai) by Germans. The modern Ice- likely from Norse habits, but actually 
landic diphthongs corresponding to corroborated by existing English uses. 

^ 1, No. 3. 



Delius.-' These divisions occur even in words wMcli in modern 
times have received the sound of (ee) or (ee), as well as in such 
words as: poiz fu ocis en traison 51,^ et en Prance mainte envaie 
135, guerpi out toz li plein pais 529, where the separation still 
remains in : trahison, envahir, pays, and the pronunciation has 
altered in the last word only. 

Aider in the I^orman war-cry is always a'ie ; 

Franceiz crient : Monjoe. e Normanz : Dex^ die. 4665 
The complete : aider, occasionally occurs, and this divided form 
seems etymologically more ancient than the diphthongal : aider, 
which is however more common.^ It is worthy of remark that the 
diphthongal pronunciation (ai'der) remained well into the xvith 
century, as it is classed with : aymant, hair, as having both vowels 
pronounced by Meigret (supra p. 118), and Ramus, 1562, classes: 
paiant gaiant aidant (Livet, p. 205). The older pronunciation of 
this one word, therefore does not admit of doubt. 

Par false e par feinte hdine 
Fu faite ceste desaisine. 


This word : haine, is now pronounced (een). Feline writes (en), 
but: hair is (a,iir) not (eer, air), haissable (a,isabr). The verb is 
now very variable: je hais, tu hais, il hait; nous haissons, vous 
haissez, ils haissent. The old French : hadir, cited by Diez, seems 
to imply the greater antiquity of the divided vowels. 

Spr. ii, 86) that the black letter v, x of 
the middle ages only differed by a small 
tail affixed to the latter, and this he 
supposes induced the scribes to abbre- 
viate the frequent termination tis, ux, 
that is, t'5, vx, as they should have been 
written, into x, which meant v with a 
subscribed x, and also led them to write 
X for V. Modern editors, he complains, 
have overlooked this, and hence written 
this pseudo x for v, in characters where 
the resemblance of form has altogether 
disappeared. So that now we find 
generally at one time als, els, Jils, at 
another ax, ex, fix, and even where 
there was no s, at one time diu, at 
another diex, or dleu, which are, Dr. 
Rapp thinks, entirely due to errors of 
writing or reading. Hence we must 
always determine in the printed copies 
whether x stands for s, u, or us. To 
this abbreviation Dr. Eapp also attri- 
butes the German proverb, to make 
one an ij; for a w, " einem ein X fiir 
ein XJ machen," that is, substitute the 
false for the true, which he thinks is a 
proof that the custom was objected to 
even in the middle ages. 

* It. aita, Pr. ahia, 0. Fr. aide ai'e, 
Fr. aide, Eng. aid, It, aitare, Pr. aidar, 
Fr. aider. Donkin's Diez's Eom. 
Die. sub ajuto. 

^ Maistre "Wace's St. Nicholas. Ein 
altfranzosisches Gedicht des zwolften 
Jahrhunderts aus Oxforder Hand- 
schriften, herausgegeben von Dr. Nico- 
laus Delius, Bonn, 1850, 8vo. pp. 95. 
" Eben so unentbehrlich erscheinen 
die Trennungspunkte liber zwei Voka- 
len, die sonst, zur Beeintrachtigung des 
Verses, fiir einen Diphthong gelesen 
wiirden, z. B. eu, di, u. s. w. Die 
Methode franzosischer Editoren im 
ersteren Falle eu, blesceure u. s. w. zu 
schreiben, ist schwerlich zu rechtfer- 
tigen, da ein so betontes e wohl kaum 
von dem folgenden Vokal verschluckt 
worden ware, wie das in der neuern 
Sprache doch geschehen ist ; eu, blessure 
u. s. w." Preface, p. xi. Dr. Delius' s 
reason may admit of dispute. The 
proper method is, of course, to follow 
the manuscript, and leave the rest to 
the reader, but in the present case I 
shall use di, e'i, as the object is to point 
out such cases to the eye. 

2 The simple figure refers to the 
verse in the Roman de Pou. The let- 
ters B, E, refer to Benoit (supra p. 438, 
note 2,) and Eustache (Roman d'Eus- 
tache le Moine, edited by F. Michel, 
Lond. 1834, 8vo). 

3 On this extraordinary form Lex 
for JDeus^ Dr. Rapp remarks (Phys. d. 


Mult veissiez .... 

Homes a terre jambeter, 

E ohevals resncs trainer. 6737-44 

The modem French is (treene). The divided vowels again appear 

to be more ancient.' 

Ausi cum glaive ist de gayne 

U cum lion prent sa rabine. B. vol. i. p. 16. 

Here again the modern French is (geen), but the divided vowels are 
more ancient.^ For ei. 

Emme sa fille fu reine 

A lie fu Engleterre encline. 6548 

The modem French is (reen), but the g extruded from regina shews 
the divided vowels to be the more ancient, and they were more 
common in this word in old Norman. Even the form : roine is 
found in "Wace's Brut. 

Grant partie sor la marine 
Malgre sa feme la rdine v. 43. 

Compare also 

Tu meisme, dist Rou, as fet ton jugement. 2029 

The following examples are curious : 

Sire, dist un Visconte, jo vos dirai ja veir, 

Cele vile n' est pas legiere a asse'ir 

Par I'ewe e par li pont povez sovent veir 

Chevaliers eserjanz cha dedenz recheveir. 4196 

Tuma sei pur li cors veir : 

Gis tei, dist il, ne te moveir. 5462 

En la boisiere volt veir, 

Ne sai s' il out de rien espeir. 5688 

Here we see a divided: veir, rhyming with an undivided: -eir. 
Now the hypothesis that ir was in such a case pronounced as eeV, 
seems contrary to all possibility or probability. But this might 
be simulated by the prefixing of an e, thus making the ordinary : 
veir into : veeir, so that in this case we should not so much have 
a divided e'ij as an omitted e. This notion is partly sustained by 

A plusors des Baronz a monstre son cimseil ; 

Si Ten tindrent trestuit a bon et afe'il, 3314 

Ki li donouent tel cunseil 

Ne li unt pas estefeeil. 8483 

where the same word feil, L. fidelis, rhyming with the same word 
cunseil is at one time spelt feil and at another feeil, which I have 
interpreted by a diaeresis. This may however have been only a 
scribal accident. Still this insertion of e is similar to the familiar 
use of u or eii as the metre seems to require. This explanation 
hardly applies to 

Normendie prendront e tendront soubs lor peiz 

E se voudrent la France partiront entr' eiz, 3633 

1 It. traino, Sp. tragin, Pr. trabi, Fr. ^ jt. guaina, Fr. gaine, 0. Fr. gai'ne, 

E. train (0. Fr. train), from trahere ; Eou. vp'aine, W. gwain a sheath ; from 
vb. It. trainare, Pr. trahinar, Fr. trainer. vagina. Milanese has guadinna, Vene- 
The suffix ino is not added to verbs, so tian guazina, Donkin's Diez. 
the Ital. and Sp. forms may have been 
borrowed from the Pr. Fr. trahim traim. 
Donkin's Diez. 

§ 1, No. 3. NORMAN FRENCH EI, AI. 457 

and it seems more natural to suppose that (e,i) and (ei) were found 
sufficient rhymes, when a trouvere was hard pressed. But what- 
ever explanation is adopted, we must remember that whereas veir 
is generally a monosyllable, it is made a dissyllable in these places 
for the exigencies of the metre, which could hardly have been done 
unless it contained within itself the elements of resolvability, by 
containing two vowel sounds usually diphthongizing. This reminds 
us of the division of ueine, mayn into ueine, mayn for the exigencies 
of the music only, and even against the metre, in the Prisoner's 
Prayer, p. 432, line 7, and p. 433, line 6, of the music, which cer- 
tainly could not have been attempted if both vowels had not been 
sounded. See also the apparent division of the diphthong in 
Chaucer, supra p. 264, and Havelok, infra p. 476. The double 
orthogTaphy : esmaier, esmaai, the last of which rhymes with ae, in : 

Guert, dist Heraut, ne t' esmaier, 

Dex nos pot bien, s'il volt, aidier. 13015 

Guert, dist 11, nos anemiz creissent ; 

Chevaliers vienent et espeissent, 

Mult part en vient, grant poor ai ; 

Unkes maiz tant ne m' esmaai. 13027 

is scarcely comprehensible on the supposition that a was not clearly 

These quotations seem to establish the existence of ei^ ai as diph- 
thongs, and as divided vowels with the pronunciations (ei, ai) and 
(e,i, a,i) and the confusion of ei, ai when ai was an undoubted diph- 
thong as in aider, compare sentreeident = s'entr'aidcnt, in the Nor- 
man version of the Proclamation of Henry III, p. 502, 1. 2. 
The question then becomes whether this pronunciation was uni- 
versal, or whether ei, ai were not occasionally pronounced (ee) as 
at present. 

Now in the first place we must not lose sight of the fact that 
several words were spelled indifferently with e or ai. 

Odes n'en volt pur li rien/ere, 

Orguil respundi e cuntrere. 6612 

Oil n'en osa plus mentfere 

Dez ke li Dus le rova tere. 7057 

Ki a sun cuer vunt a cuntraire 

Maiz n'en pot il a eel iemsfaire. 8433 

E de la grant destrucion 

Ke paen a Dol ox&a.tfet : 

S'il en France venir les lait. 6946 

Se il nel fet, a nul jur mais 

N'ara trieves de li ne pais. 8453 

Mez par li bons clers ki Tescristrent. 37 ' 

Ne mez tant com Ten vait disant. 59 

Sul Deus est sachanz e mestre 

D'Occean fist eissir e naistre B. vol. i. p. 5 

Compare : estre maistre ib. p. 10. If we examine old French, 
as distinct from Norman, we shall find the interchange of ai, e con- 
stant. It is almost impossible to open Eoquefort's Dictionnaire at 
hazard without finding examples. But at this early period, xn th 
or xin th century, I have not yet seen the confusion in many words. 
In the Roman de Eou, the only final words in -ere for -aire which 


I have noticed are : fere, tere, contrere, and these, so far as I have 
obsers'cd, do not rhyme with words that are not also spelled with 
ai. Such words would, therefore, be probably words of double 
sounds, and if we met a rhyme like : faire cuntrere, we should 
naturally suppose that the scribe had mistaken in spelling one of 
the words. Thus, in the lines just cited, for : fet lait, read fait 
lait. This is precisely similar to the double forms in Chaucer : 
dye deye, ye eye, etc. (supra p. 284-6.) That the change had taken 
place in a large number of Avords in the xiv th centuiy we see by 
such English words as : ese, pees, cler = aise, paix, clair, in 
Chaucer, but the double form : ese eyse, shew that the tradition 
at least, of the old diphthongal form was not lost in England (p. 265). 
In this examination it would be necessary for certainty to revert to 
original manuscripts of a known date, for at a late period scribes 
must necessarily have confused spellings which had come to be 
identical in signification. 

The Normans, if they carried with them IS'orse traditions, as in- 
terpreted by modern Icelandic,^ into the French pronunciation, 
must have had a tendency to palatisation ; they must have been 
fond, that is, of prefixing or subjoining i to any other vowel, either 
always or occasionally. This is fully borne out in the Eoman de 
Kou. Thus, for preceding e: triege 1362, trieves 1320, legiere 
1323, aidier 13016, chierte 1571, cunquiere 4677, similarly matiere, 
baniere, chief; mangie, eslaissie, E. p. 4, the practice being common. 
Eor a succeeding i we have the frequent termination -aige co-exist- 
ing with -age, as langage usaige 5217, messaige passaige 10790, 
rivaiges damaiges^ 127, and : tuit = tout, tons 1074, trestuit == 
tres tons 1076, where the change is made to rhyme with : s'enfuit, 
muit, deduit, but all the forms : tuit, tut, tot, are found. ISTow to 
this Korman tendency I attribute the addition of an i to a pure e, 
as in dei=^e 3770, creimon 14966, compare cremuz 15049, and such 
common forms as: sei mei tei dei mescrei lei porkei 2021-8, meiz 
3636, which are all alterations of a Latin e in the direction of pala- 
talisation, whereas the French forms : soi moi toi etc. = (sue mue 
tue) etc. are in the opposite direction of labialisation.^ Compare 
also: vezins 186, with: veizin 2292, which seem to show how 
Latin i passed through IMTorman e before it became N^orman ei, as a 
palatalisation of the e. From insufficient research I have not met 
with -tei for -te, answering to the Latin -tas, but Mr. Payne says he 
has found in Lymage del Monde, Harl. MS. 4333, dated 1246, all 
the forms : pouretei humilitei ueritei, vanitei, vanite, and similar 

1 See an account of Icelandic, infra. il faut noter que 1' accent Valeriguais 
§ 4, No. 2. See also supra p. 454. differe sensiblement de 1' accent cau- 

2 In addition to the observations at chais ; a St. Valery on ferme les 
the close of the note on p. 120, M. lettres : a devient e, et e devient i. 
r abbe Delalonde, (p. 438, n. 3), says : Je n'ai jamais entendu dire rivaiges 
"La pronunciation n>a?V7e, etc., n'existe 3 gge supra p. 131, note, col. 1; p. 
pas dans la Seine-Inferieure, sauf a 138, note, col. 1; and p. 187. A lady 
St. Valery-en-Caux, oil Ton pourrait informs me that (sue, mue, tue,) etc., 
trouver quelque chose d' analogue : on were the received pronunciations, when 
dirait plutot a St. Valery rivege : mais she was in French Canada. 

§ 1, No. 3. NORMAN FRENCH EI, AI. 459 

varieties in tlie past participle. I am inclined to class these forms 
with, the others as JS^orman palatalisations, but of less frequent 
occurrence than those with which we are so familiar, and confined 
to particular writers and localities. 

This discussion is necessarily left in a very incomplete form, and 
it is evident that lengthy researches would be necessary to arrive at 
a satisfactory conclusion, ]!^evertheless, it seems to me, that a high 
degree of probability has been attained for the theory that when the 
scribe wrote ei, ai he meant (ei, ai), or (eei, aai).^ The true English 
diphthongs were derived from the Saxon, eg, ag, ceg, and passed 
through (ejh, ajh, sejh) most probably, to become finally fused into 
(ai). They do not in any respect depend upon the I^orman, and 
hence, from the rhyming of : awey wey, both from ags. weg, and 
hence both necessarily (wai), with the IS^orman: crei Tripolay, in 
the passage which has led to this discussion, (supra p. 449), we 
should conclude that the Anglo-JN'ormans said (-ai) rather than allow 
the unproved theory that the Angio-Kormans of the xiii th century 
called : crei Tripolay (kree Tripolee), to establish by a single ex- 
ample the English pronunciation of: awey wey, as (awce* wee), in 
contradiction to the evidence that the diphthongal (awai* wai) were 
recognised by Dr. Gill as late as 1621, and still exist dialectically. 
Such a conclusion would be similar to the theory which, starting 
from modern use, makes old English long i == (oi), finds the same 
sound in Anglosaxon, and even imagines that the old ISTorman was 
pronounced so in England, so that the rhymes : cp^c fire dire shire 
of our song (p. 449) should be : (soir fair doir shoir), an hypothesis 
which our examination of long i in the xiv th century (pp. 270-297) 
must render extremely improbable.^ 

1 Mr. Joseph Payne, as a conse- ing the northern habit of (ee) to have 

quence of his researches on Norman co-existed from, at least, the beginning 

orthography, etc. (supra p. 438, note of the xvi th century in Scotland, supra, 

1), dissents from the conclusions in the p. 410, note 3, and perhaps at a still 

text respecting the Norman value of earlier period in some districts of Eng- 

ei, ai, which he believes to have always land, probably north-midland, supra p. 

had the sound (ee), and he considers 452, note, col. 2, although even there it 

that the French rhymes cited supra p. is unlikely that the forms (ei, ai) had 

264 would tend to prove that Chaucer invariably the sound of (ee). See also 

also pronounced his ei, ai as (ee). So infra p. 473, note 1. I much regret 

far as I understood, he considers that that owing to Mr. Pajiie's researches 

ei, ai had the same sound (ee) from not being yet (April, 1869) in type, 

the earliest times in England, but I am unable to exajnine the proofs 

that ai, ei had the sound (ai) in the which he has adduced, but no one can 

English of the xvi th century, as hereafter properly appreciate the evi- 

well as that of (ee) which Hart accepts dence on which a decision has to be 

as the only sound, supra p. 122. The taken, without thoroughly examining 

reader is referred to pp. 118-124, what he has so carefully and con- 

p. 238, pp. 263-266, to the rhyme ay, scientiously adduced. 
mei=English aye, Latin mei, p. 447, ^ Nevertheless as M. Le Hericher 

and to the use of a:^^, 033 in Orrmin, has advanced an opinion that the pro- 

infra p. 489, as well as to the preceding nunciation ai (ai) for long i was by no 

investigation, for the reasons which lead means unknown to the old Norman 

me to the conclusion that ei, ai were (ei, language, and has stated that it is even 

ai), or simply (ai) from the earliest times known in the modern Norman dialect, 

to the end of the xvi th century, allow- it is necessary to consider what he has 



Chap. V. 

Our knowledge of English pronunciation in the xm th and xiv th 
centuries, is now so much more certain than any knowledge which 

advanced. The following are the words 
of his assertion, Histoire et Glossaire 
du Normand de I' Anglais, etc., i. 27, 
"On retrouve en Normandie 1' I 
ouvert des Anglais, c'est-a-dire A'i. 
Dans la Hague on dit : " II est ea 
praison;" c'est-a-dire prison, "ilest 
jolai," c'est-u-dire joli. Ce son d'ail- 
leurs n'etait pas etranger au vieux 
normand, comme le prouve ce vers de 

Eve est isle, Zornee (thorn) est es- 
paine (epine) 

Soit rain, soit arbre, soit raine. 
Les paysans de Moliere, c'est-a-dire de 
rile-de-France, prononcent quelquefois 
ainsi ; voyez dans Facte II de Dotn 
Juan: 'Chagraine, Chopaine.' Mais 
les exemples sont assez nombreux en 
vieux normand ; outre celui de Wace 
nous pouvous en citer un de Beneois : 

Noise, meslee n'atame, 

Gardez que chascun en devine. 
Nous pouvons encore en citer un mo- 
derne, tire d'une chanson patoise, sur 
le nom propre Edeline : 

Vous y v'la done, monsieur Edlaine. 

( Condoleanee haguaise^ par Edeline.) 
Le paysan bas-normand rentre dans 
la prononciation anglaise de I'Y final, 
par exemple To sanctify, lorsqu'il 
dit "Tu betifai'es," tu dis ou fais 
des betises; et il prononce Envaie, 
envie, comme 1' Anglais prononce Vie, 
apocope du mot normand. Du reste, 
c'est aussi la prononciation de Picardie, 
oil le mot " Arnould daine" est devenu 
celebre. Le normand a traduit en ei 
ri du latin, que le fran^ais a traduit 
enoe: i)a;f (digitus), Freid (frigidus), 
Peil (pilus), Neir (niger), Feis (pisus), 
Sei (sitis). C'est ainsi que la forme 
primitive Franceis, Angleis, Baneis re- 
presente Francis Angli, Baniy We 
have seen the uncritical manner in 
which this author cites Palsgrave, 
supra p. 120, note, making him assert 
that in the French of his time A was 
pronounced as the modem French a, at, 
whereas Palsgrave gives a as the gene- 
ral sound, and a'i not ai, that is (ai) 
not (ee), as a sound of « in a very 
limited class of words. I therefore 
considered it necessary to check the 
assertions in the above quotation as 
well as I could. My friend Mr. W. 
Babington, being resident at Havre 

when this passage came under my con- 
sideration, obligingly made inquiries 
for me of the vicars of Notre Dame at 
Havre, Messrs. Herval and Le Due, 
and of Norman gentlemen from the 
different departments of Seine In- 
ferieure, Calvados, Orne and Eure, but 
could find no trace of this pronuncia- 
tion of long i as a'i (ai). M. 1' abbe 
Delalonde (supra, p. 438 n. 3) whom I 
also consulted on this point, writes to 
me : " / change en a'i est tout a fait 
etranger a notre contree." But re- 
specting "Arnould daine," he says: 
" Le celebre proverbe est totalement 
inconnu chez nous ; il signifie bien : 
Arnould dine, .... quant a la ma- 
niere de prononcer le mot diner, je le 
representerais plutot ainsi : deinner, et 
cette prononciation est fort repandue 
parmi les paysans." This probably 
means (dEEne). As, however, none of 
these inquiries had extended to the 
precise district pointed out by M. Le 
Hericher as that in which ai was said 
for long i, viz. la Hague, the penin- 
sula containing Cherbourg, I wrote to 
M. Totain, the cure of Beaumont, the 
nearest town to Cape de la Hague, and 
he has fav oured me with the following 
reply : " Etranger au pays de la Hague 
que je n'habite que depuis quelques 
annees, je ne suis pas autant au courant 
que beaucoup d'autres de la prononcia- 
tion des habitans. J'ai cependant in- 
terroge quelques personnes de la locahte 
que j'habite, et elles m'ont affirme que, 
dans le canton de Beaumont, nulla 
part on ne dit : praison pour prison, ni 
joldi pour joli ; ni : tu betifdies pour 
betifies. On dit : il est parti en pri- 
son ; il est joli — tu dis ou tu fais des 
betises. On ne dit pas non plus envate 
pour envie." In a subsequent com- 
munication, M. Totain says : " Mon 
Maire, M. Le Taillis, Docteur m^decin, 
originaire de Montebourg," a small 
town fifteen miles S.S.E. of Cherbourg 
on the same peninsula, "m'a affirme 
que la prononciation : il est jolai, il 
est en praison, tu betifais, qui n'est 
pas usite dans la Hague, Test tres 
generalement parmi les habitants de 
Montebourg et des environs." (supra 
p. 297, note.) After this examination 
we may feel certain that the pronun- 
ciation of long i as (ai) adduced by M, 

§ 1. No. 3. 



we possess of the old l^orman pronunciation, that, as it is in general 
derived from independent sources, we are rather justified in revers- 
ing the process of investigation and using rhymes of English and 

Le Hericher is a remarkaWy circum- 
scribed local pronunciation of no his- 
torical value, although it has the pho- 
netic importance of shewing that the 
change of (ii) to (ai) is not confined 
to England, Germany, and Holland, 
but has an analogue, confined indeed 
to a very small district, but still ex- 
istent in Normandy. "We proceed then 
at once to what bears more directly on 
our present investigation, an examina- 
tion of the evidence on which he attri- 
butes this pronunciation to the old 
Norman of the xii th century. M. Le 
Hericher does not give the reference 
to Wace and it was not without con- 
siderable difficulty that I discovered 
the passage he apparently meant to 
cite in Roman de Kou, vol. ii, p. 105, 
V. 10659. Wace is explaining the 
meaning of the English word Zonec as 
he writes it, that is, Thorney^ Thorn 
island, on which Westminster Abbey 
was built, and says — not what M. Le 
Hericher has written, but — 
Ee est isle, zon est espine, 
Seit rainz, seit arbre, seit racine. 
All trace of an di — (ai) sound here dis- 
appears. The next passage cited from 
Beneois (Benoit ?) again without any 
reference, I have been unable to verify, 
but supposing that it is correctly cited 
— a very hazardous supposition, after 
the above misquotation^ — the metre re- 
quires the separation of the syllables 
a-ta-i-ne, and the rhyme becomes re- 
gular. Roquefort gives the verb under 
the forms : atainer, ataigner, atayner, 
athir, atiner =-«?«>•(?, referring to the 
low Breton atayua, and the substantive 
in the forms : atahin, ataiue, ataiuement, 
atayne, atenes, athaine, athine, atie, 
atine, attaine, attine = haine. The word 
was evidently pronounced in a variety 
of ways, and it is not an example which 
establishes anything. From M. Le 
Hericher' s assertion with which he in- 
troduces this instance, that there are 
"numerous" examples of the rhyming 
of ai with i in old Norman, it would 
seem that he had confused the diph- 
thong (ai) with the divided vowels (a, 
i), and that when, as is quite right, 
proper, and consistent, (a,i) rhymes 
"with (i), he concluded that (ai) rhymes 

with (i), which is perfectly different. 
Certainly no one who can confuse the 
two cases, is competent to make use of 
rhymes to determine pronunciation. 
We may therefore dismiss M. Le Heri- 
cher' s assertion that the pronunciation 
ai (ai) for long i was not unknown to 
the old Normand, as perfectly destitute 
of foundation, neither of his examples 
bearing in the least upon it, and both 
discrediting his method of research. 
My own examination of all the rhymes 
in Wace's Roman de Ron has not pro- 
duced a single instance of this mon- 
strosity. In the modern example from 
La Hague, as the author writes Edlaine 
and not Edlaine, this does not seem to 
be a case in point, but appears to refer 
to some other dialectic tendency similar 
to that cited by M. Delalonde of dcinner 
for diner. I have not been able to see 
or hear of a copy of the poem Condo- 
leance Haguaise cited by M. Le Heri- 
cher . Respecting the two words cited 
from Don Juan, we must remember that 
Moliere lived in the xviith century, 
hence his ai, not di, should mean 
(ee). There are many curious spellings 
in Le Festin de Pierre, Act 2, sc. 1, as ai 
for oi and conversely, ar for er, i for 
u, but perhaps no cases of ai for i except 
those cited : " Iglia que tu me cha- 
graines V esprit, franchement." " Je 
m'en vais boire chopaine pour me re- 
bouter taut soit peu de la fatigue que 
j'aie eue." The esprit, fatigue shew 
that there was no general change. M. 
Totain says in reference to words in 
-ine, as " poitrine, chagrine, vermine, 
chopine, etc., nos paysans les pronon- 
cent generalement comme s'il y avait : 
ene ou aine. Ainsi ils disent ; Yiens 
here une chopene ou une chopaine, 
c'est-a-dire ; Viens boire une chopine." 
This confirms the above view of Edlaine. 
After this examination it would be un- 
safe to build upon M. Le Hericher' s ac- 
count of Norman pronunciation, which 
begins with an assertion very far from 
being borne out by his subsequent re- 
marks, even supposing them correct : 
" Quand la prononciation normande 
n'existera plus, on pourra la retrouver 
presquetout entiere, dans la prononcia- 
tion anglaise." — Credat Judceus I 


Norman to elicit the English pronunciation of Norman. Of course 
it is necessary to be sure that apparent rhymes are meant to be 
such, and to exclude assonances when consonants are to be deter- 
mined, and not to deduce anything from single instances, which 
may be only scribal errors. For example the passage last cited 
(p. 449) could not be used to deduce the pronunciation of any of the 
Norman words, except : tere, sarmoun, which certainly rhyme with : 
were, adoun, in the last stanza, and which must therefore have been 
called (tee're, sarmuun*), an important conclusion as respects the 
last word, as it excludes the idea of the English having heard any 
approach to the modern French nasaKty in the last word. It is 
evident that in the former part of the stanza the Norman words 
may rhyme with Norman and the English with English throughout, 
as shewn by the italics for the Norman in : defere sovent, faire 
shent, Engletere gent, faire parlement ; eyre crey, fire awey, dire 
Tripolay, shire wey, and hence no information would result. The 
construction of ballads is so loose that we have really no right to 
assume anything else, if we take the middle rhymes into account. 

The following lines are curious (Pol. Songs, p. 49, from Harl. 
MS. 978, undoubtedly of the xiiith century, supra p. 420, n. 1). 

Competenter per Robert, robbur^ designatur : 
Et per Riehard riche hard congrue notatur ; 
Gilebert non sine re gilur appellatur ; 
Gefrei, si rem tangimus, mjofrai commutatur. 

The consonants must here not be pressed too hard, and we cannot 
be certain that Robert was pronounced Roher as at present. The 
Gilelertj gilur = Gilbert guiler, shew the identity of Norman and 
English i long, guaranteed as («V, ii) by the present and perhaps 
ancient short vowel in the first syllable of Gilbert ; and Gefrei, jo 
frai = je ferai, is useful in assigning the pronunciation of Geoffrey 
as (Dzhef'rai*). But (Dzhef'ree*) must have also been in use, see 
p. 498. There is scarcely anything else which is useful in the 
Pol. Songs, but the following may be noted, the French words 
being italicised as before : pas was p. 189, De be p. 191, Boloyne 
moyne assoygne loyne Coloyne Sesoyne p. 191, Dee contree p. 216, 
eglise wise p. 251, and the Latin: custodi modj p. 251. 

There are three poems from Univ. Camb. MS. Gg. 4, 27, in which 
many French rhymes occur.^ This MS., from which also the Chaucer 
Society are printing the Canterbury Tales, is supposed to belong to 
the first half of the xv th century, but evidently cannot belong to a 
Southern locality on account of its treatment of the final e.^ Although 

^ In the spelling robbur, gilur the « ^ See an interesting account of this 

stands for e as usual ; the English MS. and its numerous peculiarities, 

reader should not think of such a sound prefixed to the Chaucer Society's re- 

as (a) or (.i). print. It may be compared with 

Audelay (supra p. 450, note 2), in the 

2 These were printed 11 July 1864 interchange of o with a, e, w, the use of 

for private circulation by Rev. H. ony for any, the frequent use of e for i, 

Bradshaw, of King's College, Cam- the neglect of final e, and in many 

bridge, to whose kindness I am in- other points, so that its authority on 

debted for the copies from which I questions of Southern pronunciation is 

quote. very slight. 


these rhymes do not properly belong to the period of this chapter, 
this seems the most appropriate place for their consideration. The 
first stanzas of the poems are as follows : 

I. De Amico ad Amicam. 

1. A celuy qui pluys eyme en Mounde 
Of alle tho that I haue founde 

Saluj od treye amour 
With grace ioye and alle honour 


2. Sachet bien pleysant et beele 
That I am ryjt in good heele 

Laus cristo 
Et moun amour done vous ay 
And also thynowene nyjt and day 

In cisto 

II. Responcio 

1 . A soun treschere et special 
Fer and ner and oueral 

In mundo 
Que soy ou salt} et gre 
"With mouth word and herte fre 


2, leo vous sanj debat 

That je wolde of myn stat i 

Sertefyes a vous ieo say 
I wil In tyme whan I may 


III. [The Songs of the Birds] 

1. In may whan euery herte is lyjt 

And flourys frosschely sprede and sprynge 
And Phebus with hise bemys bryjte 
Was in the bole so cler schynynge 
That sesyn in a morwenynge 

Myn sor for syghte to don socour 
With inne a wode was myn walkynge 

Pur moy ouhter hors de dolour 

2. And in an erber sote and grene 

That benchede was with clourys newe • 

A doun I sat me to bemene 

For verray seyk ful pale of hewe 
And say be syde aturtil trewe 

For leue gan syngyn of hire fere 
In frensch ho so the roundele knewe 

Amour me fait souent pensere. 

The following arrangement of these rhymes will shew their 
hearing. The French words are in Italics, the references to the 
number of the poem, as above, and the line, explanations in 
brackets : 
A. debat senbat [s'en bat] iii 22, dehat E. le [U, broad] me i 52, le the ii 28 — 

stat ii 7 — special oueral ii 1 — alias pete [pite] me ii 40, verite the i. 23, 

was ii 31 — toward gard [garde] i 70 charitehe i 67, volunte the [thee] i 37, 

AI. ap [ai] day i lo, serray [serai] ii 46— ^re [gre] fre [free] ii 4, tre- 

day ii 13, say [sais] may ii 10 same [tres aime] be i 55, tresame the 



Chap. V. 

stedefastly [another faulty northern 
or XV th century rhyme] ii 52 — 
fere [ = fyr = fire for this rhyme, 
see p. 272] aymyer iii 38, quoer 
[coeur] fyr [evidently taken as (kcer, 
leer), see last case] i 40, entyre de- 
parter [compare the last case but 
one] iii 118 — dy) [dis] pris i 31 — 
tryst [triste] nyjt [night, see re- 
marks below] i 19. 

0. a cestyn ay maunde de vous ore [or?] 
more ii 43, note rote i 46, sort mart 
iii 62. 

OU. verteuous ioyous [joyeux] iii 86, 
amour flour ii 22, amour honour i 4, 
socour dolour [douleur] iii 6. 

Nasals. — penaunce languissaunce iii 
70 — dolent schent ii 19, entendement 
entent i 58, greuousement schent ii 
37 — seyn [sain] serte)Ti i 49 — 
— mounde [monde] founde i 1. 

[thee] i 13, done [donne] the i 61, 

en presone [emprisonne] sle [slay, as 

often in Chaucer] i 34 — fere [com- 
panion] jo^'ws^r^ [penser] iii li^manere 

were ii 34, chere pere [peer] i 43, 

et pur ceo leo vous creser [?] daunger 

i 28,— lea I [loyal] fel [feel] i 16, beele 

[belle] heele [health] i 7. 
EI. weye soye [sois] iii 46, espeye 

[epee, should be espie, the e was a 

subsequent insertion] deye [should 

be dye as often in Chaucer, p. 284] 

EU. re we adetve iii 94, crew deceu iii 54 
I. vye [vie] curteysye ii 49, pry [prie] 

curteysy [should be curteysye as in 

the last case] i 64, ermony [should 

be harmonyel oublye iii 30, maladye 

sikyrlye [should be sikyrly, but 

then the rhyme is faulty in a 

northern or late xv th century man- 
ner] ii 16, ieo vous pry [for prye"] 

So far as these rhymes establish anything they go to confirm our 
former conclusions in every respect, and to shew an absence of 
nasality in the English pronunciation of French in the xv th century, 
as we shall find again in the xvith, Chap. YIII, § 3. The rhyme : 
tryst nyjt, is very remarkable. It cannot be supposed either that 
^ was in such a position as ny)t ever pronounced as 5, although we 
find dy) = dis i 31 in the French; nor on the other hand can we 
suppose that s was omitted in tryst and ^ in ny^t, producing the 
rhyme: (triit, niit,) because s is still pronounced in this French 
word. Hence we are compelled to assume an assonance (trist, 
niA'ht), which a clumsy poet found quite near enough to satisfy his 
ear. JVLr. Lumby however entertains a different opinion. In his 
edition of King Horn, infra p. 480, n. 1, from this same Cambridge 
IMS. Gg. 4, 27, 2, he observes on the forms, mifte = mijte 10, 
dofter = dorter 249, rhyming with fo^te, and rift ^ rijt in 
line 663 of Flori) in the same IMS., which line also contains no jt, 
with^ and not f : " This interchange," he says, '' occurs so often in 
early ]MSS. that it is a conclusive proof of a similarity in sound be- 
tween the letters," and adds that " in several copies of Piers Plow- 
man soure occurs for ^oure,^^ ^ and refers to Rel. Ant. i, 48, for a poem 
where this substitution occurs throughout. This poem. The Five Joys 
of the Yirgin, is from Trin. Coll. IMS. E 14, 39,^ which Mr. T. Wright 

1 Mr. Skeat knows only of one copy, 
MS. Cotton Yesp. B. xvi, where there 
are several, but not many, examples, 
and the spelling is altogether singular. 

2 Some account of this MS. is given 
in Mr. Albert Way's Preface to the 
Promptorium Pavvulorum, p. Ixxii, 
under the heading " Femina." This 
MS., I am informed by Mr. Aldis 
Wright, the librarian of Trinity Col- 
lege, disappeared from that library 

between 1853 and 1859, and as no one 
had taken it out on bond in that inter- 
val, it must have been appropriated. 
There are notices of it in Hickes, The- 
saurus i, 144, 154, and its disappearance 
is a serious loss to Early English phi- 
lology. The poem of the Five Joys 
is reprinted in Golbeck and Matz- 
ner's Sprachproben p. 51, but these 
editors have taken the liberty of replac- 
ing -ft by -ht throughout. 


assigns to tlie first half of the xni th century, a conclusion at 
variance with the orthography thou which is invariable and occurs 
frequently, and wid-oute. The only other test word is ure, which 
has the xiii th century form, so that the close of the xin th century 
is the time indicated, as for Havelok. The words containing f for j 
in this poem are : brift mift, lifte rifte, mifte, drift rift, mifte 
brifte, brift. This same poem contains some other curious ortho- 
graphies as : sue [such], seal, sculde, scene. It omits the guttural 
altogether in : broutest [broughtest], slo [slew]. It apparently 
confuses v with \ in 

The thridde dai he ros to hve ; 
Levedi, ofte were thou blive [blij^e ?] ^ 

Ac never so thou were tho. 
Levedi, for then ilke sive [si be ?] 
That tou were of thi sone blive [blijjc ?] 

Al mi sunnes thou do me fro ! 

In the last stanza we have : bene newe, printed, meaning ap- 
parently : bene newe, which would be an assonance, and is the 
reading adopted by Matzner. 

Levedi, tuet thou me mi bene 
For the joie that ever is newe. 
Thou let me never be furlorn. 

These peculiarities render this text not particularly useful for our 
purpose, and inasmuch ' as ) was used for both % and 5, some inac- 
curate scribes may have considered that f , which was also certainly 
(z) at times, might be used for 5. The only passage I have yet met 
in which ^ standing for 5 has apparently the sound (s), is this very 
suspicious couplet of a poem full of bad spelling (i 19, supra p. 463) : 

Jeo suy pour toy dolant et tryst 
Ther me peynyst bothe day and nyjt 

and it would be unwise to found a theory upon a single instance of 
such small authority. In the first passage of King Horn, the 
parallel MSS. in Mr. Lumby's preface, p. vi, give myhte, 
m i c t e ; and m i j t e occurs two line above in his own text. 

These rhymes of Korman and English are rather to be treated as 
jokes than as serious attempts to determine the Norman pronun- 
ciation. They may be classed with Hood's description of an Eng- 
lishman's difficulties in Erance : 

Chaises stand for chairs, For wine I reel'd about 

They christen letters Billies, To show my meaning fully, 

They call their mothers mares, And made a pair of horns 

And all their daughters ^//^>« ; To ask for " beef and bully ^ 

Strange it was to hear, Then their cash was strange, 

I'll tell you what's a good 'un. It bored me every minute. 

They call their leather queer, How here's a hog to change, 

And half their shoes are wooden. How many soivs are in it ! 

Comic Annual, 1831, p. 82* 

^ Blive means quickly, which will line biliue stighe (Prisoner's Prayer 

not make sense here. The rhyme here 27), because (f, th) and therefore (v, 

then sinks into an assonance, which dh) are more readily confounded than 

even more resembles a rhyme than : (v, ^h) ; we may suppose bli]>e to have 



Moore's Fudge Family in Paris, shews : joy Roi, St. Denis penny, 
swear is Veri/s, throat papillote, fond Fronde, eracker ^acre, Natties 
pdtes, affiches wish, Musses use, mon Prince sense, jolie Dolly, 
icrevisses bliss, coach poche. In Byron we find : true is petits putts 
(Juan, 15, 68) eprouveuse muse (ib. 9, 84), Vauhan hang slang (ib. 
5, 11), a V Allemande understand hand (ib. 15, 66), French Per- 
venche 14, 75. These modem instances should teach us not to ride 
our old examples too hard, and certainly not to draw conclusions 
from a few cases. 

4. The Story of Genesis and Exodus, cmcX a.d. 1290. 

Mr. Richard Morris attributes the composition of the rhymed 
account of Genesis and Exodus contained in a MS. in Corpus Christi 
College, Cambridge, to about a.d. 1250, but the actual writing of 
the MS. to nearly as late as 1300,^ so that it was ''the work of a 
scribe to whom the language was more or less archaic." The 
dialect he considers, together with that of the Bestiary (supra, 
p. 439), and the Orrmulum (infra, p. 486) to be East Midland. 
This poem being well known to all the members of the Early 
English Text Society, I have examined the rhymes to obtain indi- 
cations of the pronunciation, and shall refer to them by the number 
of the lines in which they occur. 

Assonances are not uncommon, but the principal are those in 
which an, corresponds to am, as: ^an nam 481, nam canahan 725, 
abram leman 781, abram iurdan 805, abram maji 909, bigan abram 
921, abraham ^an 1189, nam laban 1653. Occasionally in im, 
caym kin 543, elim sin 3307 ; on om, on-on horn 2199 ; un um, 
cumen munen 1621. Probably : gate quake 1054 is an error of the 
scribe for : gate quate. Joseph swep 2085, hond wrong 2063, 
sokoth pharaofh 3209, are single cases, but oe occurs more fre- 
quently : fot oc 2497, oc mod 3923, mo^ boc 3603. Altogether 
false rhymes are rare, and are probably scribal errors : agen under- 
gon 1159, drog nuge 1327, get bigat 2277, ^or ger 2417, specande 
lockende 2821, moysen man 3109, eliazar or 4091. In: numen 
comen 343, broken luken 361, 3779, this is almost certainly the 
case, and in: swem greim 391, which would otherwise be an 
example of e, ei rhyming, the second word should be grem or grim. 
The rhyme i, e, is normal, as in Chaucer, (supra p. 272) : K^er 
ne^er 369, effraym hem 2151, wliten eten 2289, abiden deden 
2483, mide dede 2963, and probably implies that i = {ii, i). Oc- 

been called, (bliidh-e), at present both ^ The story of Genesis and Exodus, 

(bbidh, bbith) are heard. Matzner an Early English Song, about 1250, 

reads blithe, sit/ie, saying : " "Wir now first edited [for the Early English 

schreiben hier blithe fiir bltve, und Text Society] from a unique MS. in 

sithe fiir sive ; da sonst die Stelle un- the library of Corpus Christi College, 

erklarbar bleibt. Dadurch tritt in live Cambridge, by Richard Morris, Lon- 

die Assonanz an die Stelle des Reims. don, 8vo. pp. xl. 224 ; a.d. 1865. 
Dass blive = bilije, beliue, quickly, nicht 
geduldet werden kaun, ist selbstver- 


casionally an e final seems omitted, or added by mistake, as : song 
amonge 699, cliild milde 985, compare: childe mild 1305. In 
many instances -e, and -en rhyme, where the editor has apparently 
changed -e into en^ though in some cases it woidd seem more 
correct to change -en into -e. 

As regards w, it had certainly generally the pronunciation (uu), 
and those rare cases in which it is replaced by ow, may be attributed 
to the more modem habits of the scribe, as the use of ou for (uu) 
seems to have commenced about the close of the xiiith century. 
Thus we find: run = speech, circumcicioun 991, town dun = 
down 2739, but: tun dun 713, iQXQm\jii}i= Pharaoh's daughter's 
name, out 2615. But the Hebrew: man hu? nu 3329, alluding 
to Ex. xvi, 15 (man huu?), what is this?, the question asked 
when the manna was first seen, as clearly points to the use of u for 
(uu) as the cuccu of the Cuckoo Song. The use of u for (yy, y), 
probably called («V, i, e) is rare, but we find w?^^w<;?m = barren, 964. 

That the unaccented syllables were occasionally pronounced in a 
slovenly manner, we collect from the rhyme : euenehe uone 331. 

Diphthongal combinations are altogether rare. 

Ea occurs, but rhymes with e, and may be always a scribal error : 
forbead dead 311, opened dead 387, red dead 401, bead dead 1059, 
ear ^ear = ^^^r ? 1089, forked dead 1329, dead red 2513. Pro- 
bably pronounced (ee') or (ee, e) in all cases. 

Oa rhymes always with a, and may have been {aa) : moal = 
speech natural 81, woa = woe eua = Eve 237, gomorra ^oa = ^a 
839, oba woa = woe 879, salmona ^oa 3893, fasga doa = ^a 4129. 

Ai, ei rhyme together, and must have both been (ai) : ay day 87, 
wei dai 1429, grei awei 1723, dai awei 2305, day wey 2721, dai 
mai 2747. In : awei deal 861, the last word is a mere scribal error 
for dai. 

The guttural g is occasionally omitted, as : ru esau 1539, where 
ru = rough. Sometimes it is merely changed into w, probably in- 
dicating (wh) or (g^^h) : noght sowt-2869. We also find initial 
gh, in ghe = she, 237, 337, 339, but ge = she 1024 possibly a 
remnant of (^h), though (j) seems to have been the sound intended. 

This examination confirms our previous conclusions as to the pro- 
nunciation of the XIII th century. 

The following is an attempt to convey a notion of how the poem 
may have been read. The text is according to the MS., the pro- 
nunciation introduces some conjectural emendations, without which 
it would have been impossible to read the text.^ 

^ In one or two points I differ from there about as in : "Sor buten noe long 

Mr. Morris, particularly in the last line fwing lie dreg 566, .vii. mone'S "Sor- 

but four, where he takes buten hunte= buten he ben 3625, hunte becomes the 

" without search, or hunting, without infinite mood, and the construction is 

delay," but by restoring ic in the pre- ie sal hunte '^or- buten, I shall hunt 

ceding line, wanted for the metre and there about, I shall endeavour to ac- 

the sense, and taking "Sor buten to mean complish it. 



Chap. V. 

Genesis and Exodus, 269-318. 

"Wifdom ^e made ilc ^ing of 

Q,wuat-fo-eu^re on heuone or her 

if wrogt. 
Ligber he fridde a dere frud, 

And he wur^e in him-feluen 

An wi^ ^at pr/dehim wex a ny^, 

^at iwel welded al his li'S ; 

^0 ne migte he non lou^rd 

^at him fulde ^hinge grauen : 

Min fligt, he seide, ic wile up- 
Min fete nor^ on heuene maken, 

And ^or ic wile litten and fen 

Al ^e 'Shinges ^e in werde ben, 

Twen heuone hil and helle dik, 
And ben min lou^^-d geuelic. 
^0 wnr^ he drake 'Sat ear was 

^0 wur^ he mire ^at ear was Hgt, 

And euerilc on ^at helden wid 

^0 warden mire, and fwart, and 

And fellen nt of heuones ligt, 
In to ^is middil walknes nigt ; 

Conjectured Pronunciation. 

Wiis'doom dhe maad ilk thiq of 

K^^'hat-s-eer• on nevn- or neer 

is r^^;okht. 
Likhtbeer' ne srid an deer'e 

And ne wurth in nimsel'ven 

And with dhat priid -im weks 

a niidh 
Dhat ii'vel weldeth al -is siidh, 
Dhoo nee mikht -ee noon loverd 

Dhat Him suld [al'e] thiq'e 

thraa'ven : 
Miin flikht, ne said, ic wil up 

Miin see'te north on nevne 

And dhoor ic wiil*e sit*n- and 

Al dhe thiq'es dhee -n werld'e 

Tween nevne nil and nel'e diik, 
And been miin loverd gee'veliik, 
Dhoo wurdh -e draak'e dhat eer 

was knikht, 
Dhoo wurdh -e mirk dhat eer 

was likht. 
And everilk oon dhat held 'en 

with Him 
Dhoo wurdh 'en mirk and swart 

and dim, 
And fel'en uut of nevnes likht, 
Intoo* dhis mid'il walk'ues nikht; 


Wisdom then made each thing of 

Whatsoever in heaven or here is 

Light-bear [Lucifer] he [God] clothed 

in precious clothing, 
And he became in himself proud, 
And with that pride in-him waxed an 

That ill ruleth all his path. 
Then not might he no lord endure, 
That for-him should [all] things control. 
My flight, he said, I will up-take, 

My seat north in heaven make, 

And there I will sit and see. 

All the things that in the world be. 

Between heaven's hil and hell's ditch, 

And be to-my lord even-like. 

Then became he dragon that ere was 

Then became he mirky that ere was 

And every one that held with him 
Then became mirky, and black, and dim, 
And fell out of heaven's light. 
In to this middle welkin's night, 



Genesis and Exodus. 
And get ne ku^e lie nogt blinne 

for to don an o^er linne. 
Ellen lie sag in paradif 
Adam and eue in mike pr«f, 
Newelike he was of er^e wrogt, 
And to ^at mirie blifTe brogt ; 
^owgte ^is q«^ead, hu ma it ben, 

Adam ben king and eue qwuen 

Of alle ^e ^inge in werlde ben. 

Hu mai it hauen, hu mai it fen, 

Of fif, of fugel, of wrim, of der. 

Of alle ^hinge ^e wunen her, 
Eumlc filing haued he gene 

Me to forge, fca^e, and fame j 

for adam ful ^us, and his wif. 

In blifle ^us leden lefteful lif ; 
for alle ^o, ^e of hem fule cumen, 

fulen ermor in bKffe wnnen, 

And we ^e ben fro henene 

fulen ^uffe one in forwe liuen ; 

Get ic wene I can a red, 

^at hem fal briugen iwel fped ; 

Conjectured Fronuneiation. 

And Jet ne kuudh*e nee nokht 

for to doon an oodh'er sin*e. 

Eest'en He saagh in paa*radiis 

Aa'dam and Eev in mik'e priis, 

!N'eu-liik* -e was of erth-e rwokht, 

And too dhat mir-ie blis-e brokht, 

Thoukht'e dhis kw^eed, huu mai 

it been, 
Aa'dam been kiq and Ee*ve 

Of al*e thiq'e dhee -n werld-e 

Huu mai ic naan, huu mai ic 

Of fis, of fuugh-el, of wirm, of 

Of al'e thiq-e dhee wuun'en neer, 
Eer'ilk* thiq navd -e geeve 

Mee to sorgh'e, scaadh and 

For Aa'dam sal dhus, and His 

In blis'e leed'en les'teful liif ; 

For alle dhee -f Hem sul*e kuu*- 

Sul'en eermoor* in blis*e wuu*- 

And wee dhe been froo nevne 

Sul'en dhus oon in sorgh'e lii'ven, 
Jet ik ween i kan a reed 
Dhat Hem sal briq'en ii-vel speed. 


And yet not could he not cease 
For to do another sin. 
Eastwards he saw in paradise, 
Adam and Eve in much honour. 
Newly he was of earth wrought, 
And to that merry bliss brought. 
Thought this evil-one, how may it be, 
Adam be king and Eve queen 
Of all things that in world be. 
How may I have, how may I see ! 
Of fish, of fowl, of worm, of beast, 

Of all things that dwell here, 
To-every thing has he given name. 
For my sorrow, scathe and shame. 
For Adam shall thus and his wife 
In bliss lead lasting-fall life. 
For all who of them shall come 
Shall evermore in bliss dwell, 
And we that be from heaven driven. 
Shall thus only in sorrow live. 
Yet I ween I know a plan 
That them shall bring evil speed. 



Chap. V. 

Genesis and Exodus. 
for gef he don ^ad god for-bead, 

^at fal hem bringen to ^o dead, 

And fal get ^is ilke dai, 

^or buten hunte if ic mai ; 

Ic wene ^at ic, and eue hife wif, 

fulen adam bilirten of hife hf. 

Ic wene ^at ic and eue 

fulen alle is bliffe dreue. 

Conjectured Pronunciation. 
For jef He doon dhat God for*- 

Dhat sal Hem briq'en too dho 

And [ic] sal jct dhis ilk*e dai 
Dhoor buut'cn nunt'c jif ik mai" 
Ik wecn'e dhat ik and Eev -is 

Sul'en Aa-dam biliir'ten of His 

Ic ween-e [to sooth] dhat ik and 

Sul'en [Aa'dam] al -is blis'e 


For if they do that-which God forbade, I ween that I, and Eve his wife, 
That shall bring them to the death. 
And [I] shall yet this same day 
There about hunt, if I may. 

Shall Adam betrick of his life, 
I ween [in sooth] that I and Eve 
Shall [for-Adam] all his bliss trouble. 

6. Havelok the Dane, circa a.d. 1290. 

Sir Frederick Madden in his edition of this poem ^ considers its 
author to have been a Lincolnshire man, and the time of composition 
between a.d. 1270 and 1290. As the romance was popular, there 
may have been many copies, and the manuscript followed by Sir F. 
Madden may not have been original. In its orthography, apart 
from its dialectic peculiarities, (which are numerous but do not here 
come into consideration, as the object is merely to determine the 
value of the letters,) it shews a transition from the customs of the 
xin th to those of the xiv th century, much more marked than in 
Genesis and Exodus. Thus ou is frequently used for (uu), ^ou 
being the common form, though \u is by no means unfrequent, 
indeed both forms occur in the same line : Grim, \ou wost ])u art 
mi thral 527, and we have \w 1316, and \o 388, where, probably, 
a final u has been accidentally omitted by the scribe. The following 

* The Ancient English Romance of 
Havelok the Dane, accompanied by the 
French Text, with an Introduction and 
Glossary by Frederick Madden, Esq., 
F.A.S., F.E.S.L., subkeeper of the 
manuscripts va. the British Museum, 
printed for the Roxburgh Club, 1828, 
4to. This edition being very scarce, a 
new one compared afresh with the MS. 
has been prepared for the Early English 
Text Society under the title : The Lay 
of Havelock the Dane : composed in 
the reign of Edward L, about a.d. 
1280, formerly edited by Sir F. Mad- 
den for the Roxburghe Club, and now 

re-edited from the unique MS. Laud 
Misc. 108, in the Bodleian Library, 
Cxford, by Rev. Walter W. Skeat, 
M.A., London, 1869. It will there- 
fore be assumed to be accessible to all 
members of that Society, and will be 
cited by the number of the verses, as 
usual. The citations originally made 
from Sir F. Madden's edition have 
been verified by Mr. Skeat's. I am much 
indebted to Mr. Skeat for many hints, and 
for kindly allowing me to make use of 
his proof sheets before publication, so 
as to enable me to insert this notice in 
its proper place. 

§ 1, No. 5. 



rhymes serve to shew the identity of the two spellings : yow now 
160, pn'soun la^arun 330, mouth suth 433, yw = you nou 453, 
nov = now you 484, bouMe^ wnden = wounden 546, unbouwden 
fu/^den 602, hw = how he was mike, hw he was strong 960, 
doun tun = town 1630, wounde grunde 1978, bowr tour 2072, 
dune croune 2656. Of course ow, ow also occur as (oou) corres- 
ponding to ags. aw, oh, and the guttural is generally lost in (w) 
after c», thus : ynowe slowe 2682. In : croud god 2338, we should 
probably read crod^ as the proper form of the past participle.^ The 
frequent occurrence of ou^ however, would lead one to suppose 
that the actual MS. must belong to the very end of the xiiith, 
if not to the beginning of the xiv th century.'^ 

Assonances are frequent, and the more marked that there is 
often no relation between the consonants which follow the iden- 
tical vowels. Thus: rym fin 21, yeme queue 182, harde crakede 
567, befe rede 694, knaue plawe 949, staredew^ ladde;* 1037, 

^ Ags. creodan (creld, crudon, cro- 
den) Ettmiiller, Lex. Anglos, p. 400. 
Nail (supra p. 166, note 1) under 
Crowd-Barrow^ quotes : *' She sent my 
motlier word by Kate, that she should 
come hither when God sent time, 
though she should be crod in a barrow. 
Letter of Margery Paston, a.d. 1477." 

2 Mr. Skeat informs me that : '* No 
other MS. of Havelok has ever been 
heard of, or known to exist : though of 
course there may have been several. If 
this is not the original, it is at any rate 
a very early copy. I do not think Sir F. 
Madden, or any other judge of writing, 
would admit it to be later than about 
1280, the probable date of the compo- 
sition. The evident age of the MS. is 
one evidence of its early composition." 
The MS. containing Havelok begins 
with lives of Saints, and Havelok was 
overlooked for years, because it does not 
begin till fo. 204. It ends on fo. 219*, 
and is immediately followed by Kyng 
Horn in the same column. This has 
all the appearance of a copy, not an 
original MS., and as we have two other 
copies of King Horn (p. 480, n. 1), we 
may some day find another of Havelok. 
Even a much later one (as in the case of 
Lajamon) would be of great service. 
It is of course impossible to date a MS. 
by the writing only, within 30 years, 
the working life-time of a single scribe. 
The orthography would lead me to 
place the actual manuscript after the 
copying of Genesis and Exodus, and 
within the variable period, say 1280 to 
1310. Probably after the last date 
ou was universally employed for (uu). 

If the reader will turn to : Seinte 
Marherete, the Meiden ant Martyr, in 
old English, first edited from the skin 
books in 1862 by Oswald Cockayne, 
M.A., and now reissued for the Early 
English Text Society, 1866, and compare 
the three versions there given, the first 
from the MS. Ecg. 17, A. xxvii., ap- 
parently written in 1230, in which no 
case of ou ~ (uu) occurs ; the second 
from MS. Harl. 2277, attributed to 
1330, in which ou is always used for 
(uu) ; and the third from the lost Cam. 
MS. (supra p. 464, n. 1 ) as printed by 
Hickes, in which, if the text is to be 
trusted, there is just a trace of « = (uu) 
— J7U 22, prisun 26, etc., dragun 44, 
ut 28, 56, )7oru 47— amidst a great 
preponderance of ow, the value of this 
sign of age in a MS. will become more 
apparent ; compare also supra, pp. 408, 
423, 439, 445, 467, and p. 481, 1. 11. 

^ " Probably miswritten for stradden 
contended." Skeat, Glossary, Sir F. 
Madden, and Garnett are of the same 
opinion. It is with great difiidence 
that I presume to doubt this correction. 
Stradden would introduce a Norse word, 
whereas the noun strout is used imme- 
diately 1039, and verb stroute in 1779, 
from ags. strudan^ strutian, and it does 
not seem likely that both words should 
coexist in the same dialect, or, if they 
did, should be used in immediate proxi- 
mity. Nor, I must confess, does con- 
tended seem to make very good sense. 
The passage relates to the game of 
" putting the stone," the point being to 
see who should throw an enormous 
stone furthest, for he whose stone was 




Chap. V. 

shop (?) hok 1101, odrat bad 1153, drawe haue 1297, fet ek 1303, 
ioye tronc 1315, maked yschaped 1646, richo chinche 1763, 2940, 
feld swerd 1824, 2634,^ seruede werewed 1914, wewd gent 2138, 
shauwe knawe 2206, grauen namen 2528, thank rang 2560,^^ bofe' 
rede 2585, bofen drowen 2659, shawe knawe 2784. 

Apai-t from these assonances there are no bad rhymes which do 
not admit of explanation. Thus: hey fri 1071, might possibly 
be : hy fri, see p. 285, but as the form hy does not occur in Have- 
lock, we should probably read: hey sley, compare 1083.* The 

even an inch before the others was to 
be held a champion : 
Hwo so mithe putten ])ore 
Biforn anoj^er, an inch or more, 
Wore ye [=he?] yung, [orj wore he 

He was for a kewpe told. 1033 

"What would then be more natural than 
for the champions and the lads to stand 
and look intently, stare, prior to the 
throw, and then make a great conten- 
tion, strout, about the best cast. This 
is what the text says as it stands : 
Al-so )7e[i"| stoden, an[d] ofte staredew 
j^e chaunpiouns, and ek the laddew. 
And he maden mikel strout 
Abouten j^e alj^erbeste but. 1037 

It would, however, be rather curious to 
say that the champions and lads stood 
and contended and made a great con- 
tention about the best throw. If we 
must alter the passage, siraden, strode 
about (Ettm. 746), would make decent 
sense, but not so good as starede^i. It 
was doubtless the apparent harshness of 
the assonance : stareden ladden, which 
led to this conjecture. In the same 
way Mr. Morris, anxious to avoid the 
assonance : harde crakede 567, proposed 
to change 

And caste J>e knaue adoun so harde, 
])at hise croune he j^er crakede 

And caste \e knaue so harde adoun, 
Jjat he crakede ))er hise croune. 
(Skeat, p. 91). Where the rhyme re- 
quires adoune as in King Horn 1487 
(Lumby's edn.) 

Fike??hildes creme 
^per iftdde adune . . . 
which is quoted in Mr. Skeat' s glos- 
sary (from MS Harl. 2253,) as : crowne 
adoune, shewing the more ancient form 
of the other version of King Horn. 
But the only alteration really required 
is : J>er he crakede, for : he )5er crakede, 
in order to preserve the e in croune. As 

to the assonance itself, it is harsh to our 
ears only. We must remember the 
constant habit of the metathesis of r, 
so that : harde crakede, may have been 
called : harde carkede, which would 
have been almost a rhyme, as : star' den 
ladden, also is. On the principle of not 
making imnecessary changes, I prefer 
accepting the reading of the MS. in 
each case as it stands, and therefore re- 
tain both : harde crakede, and stareden 
ladden, as assonances. 

^ And ]?e ])redde so sore he slow, 
iTat he made up-on the feld 
His left arm fleye, with the swerd. 
On which Mr. Skeat remarks : " Of. 
1. 1825. We should otherwise be 
tempted to read sheld; especially as 
the shield is more appropriate to the 
left arm." This was Sir F. Madden' s 
original suggestion. But with may 
denote the instrument: he slow \)e 
))redde so sore with the swerd, ]jat he 
made, etc. Compare the constructions, 
supra p. 376, art. 110. Compare also 
the parallel passage : 
For his sword he hof up heye, 
And ])Q hand he dide of fleye. 
That he sraot him with so sore. 2750 
I feel doubtful whether the other inter- 
pretation : that he made his left arm 
together with the sword, fall on the 
field, could be justified by parallel 

"^ This may be a rhyme, see supra 
p. 192. 

3 As we find: rede be)>e 694, bejje 
rede 1680, we should of course read: 
be]7e rede in this place. This is only 
one of the numerous instances of the 
interchange of ^, a, o, to be noticed 
presently. Thus we have: baj^e 1336, 
2543, and bo>en 173, 697, 958. 

* According to the text Godrich hears 
the knights talk of Havel ok : 

Hw he was strowg man and hey, 

Hw he was strong and ek fri, 1071, 
and then he thought that King Athel- 

$ 1, No. 5. 




rhyme : yhe se 1984, is a mere misprint in Sir P. Madden's edition, 
corrected by Mr. Skeat to : Jhe se, where the h is an idle insertion, 
compare ]?e = thigh 1950, and : ^hinge = ^inge. Gen. and Ex. 300. 
The passages which present the greatest difficulty are the follow- 
ing : eir tother 410, misdede leyde 994, deled wosseyled 1736. 
The last is explained by : wesseylew todeyle 2098, which ought 
to shew that the writer had two ways of pronouncing: delen, 
deylen, (deel'en, dail'en). Compare : 

So ))at ]7e blod ran of his fleys, 
pat tendre was, and swi]?e neys, 
And woundede him rith in \e flesh 
]?at tendre was, and swij^e nesh. 

As the dialact of Havelok shews a Scandinavian character in 
many words, the form deylen may have arisen from that source, 
Icelandic at deila, (d^<9i'la) to divide, and it would be in fact more 
difficult to acccount for the forms Jleys neys} If we do not accept 



wald had made him swear to give his 
daughter to the " hexte " = highest, 
tallest, man alive, and then asks 
Hwere mithe i finden ani so hey 
So hauelok is, or so sley ? 1083 

It is evident that the two couplets 
ought to correspond. Sley, of course, 
means skilful, Havelocks skill: hw he 
warp l»e ston Ouer ]>& laddes euerilkon 
1061, having made him the common 
talk. Fri yields no good sense. 

^ Yox Jleys see supra pp. 265, 441, 
445. The form is, in fact, not unusual. 
For neys there seems to be no authority, 
and cognate languages do not exhibit 
the diphthong (ei), as they do in the 
case of high German ^ewc/j, theil, weich 
(flaish, tail, bhaiArh), compare Dutch, 
vleesck, deel^ week (vlees, deel, bheek). 
These undoubted correspondences of (e, 
ai) in high and low German, and the 
occasional use of ei in Icelandic as deila, 
veikr (dml*a, vmkr), but its rejection in 
other cases, asjlesk (flesk), may at least 
serve to render intelligible some doubt- 
ful usages in such a provincial region 
and early time as that which gives us 
the rhyme of Havelok. Not only 
does provincial, but even metropolitan 
usage at the present day, furnish 
examples which may give as much 
trouble to a future investigator. Com- 
pare the example Chap. XI. § 3, where 
it will be seen that Mr. Melville BeU 
writes : (deiz, weisted, fein, geiv, keim, 
8«i), where I have (de^z, wrested, feen, 
geeY, keem., see) = days, wasted, fain, 
ffove, came, say, though we are both 
supposed to speak the same dialect. 
See also p. 450 n. 2, and p. 459, n. 1, 

and the forms sede saide, p. 446. . . . 
After the preceding observations had 
gone to press, I received a remarkable 
confirmation of the views there ex- 
pressed concerning the possibility of 
different pronunciations coexisting in 
limited districts, from an account of the 
present pronunciation of English in 
the Peak of Derbyshire, orally com- 
municated to me by a native of the dis- 
trict, Mr. Thomas Hallara, of Man- 
chester. A somewhat detailed account 
of these remarkable pronunciations will 
be given below, Chap. XI. § 4, but it is 
as well to notice here, that on ihe west 
of the mountain ridge of the peak we 
find (mee, dee, ■ewee-, pee) and on the 
east (mii, dii, ■ewii, pii) for may, day, 
away, pay, and again on the west we 
have (shiip, ship, mn) and on the east 
(sheip, sleip, mei) for sheep, sleep, me. 
This characteristic diphthong (ei), 
found also in the west of the ridge in 
(dzheist, dzheint, beil, peint, eint'- 
mynt) for Joist, Joint, boil, point, oint- 
ment, is, as pronounced to me by Mr. 
Hallam, a sound which one Southerner 
vdll hear as (ee) and another as (ai). 
Com])SLre po?/nte =peynte, p. 447, 1. 14. 
"We can guess how a peasant of the 
Peak, with his partial inoculation into 
the mysteries of modern orthography 
is likely to write, but to put ourselves 
into the position of the most careful of 
ancient scribes, we have only to en- 
deavour to appreciate such sounds and 
attempt to commit them to paper, after 
a careful study of phonetics. The ex- 
treme difficulty of appreciation, the 
readiness with which we mentally as- 


the form deyle, then one of three things must be the case : 1 ) The 
rhyme may be faulty, but it would be perhaps the only faulty 
rhyme. Or, 2) the ey, e may be a true rhyme, but then, indepen- 
dently of previous investigations, the persistent avoidance of such 
rhymes is remarkable, and there would have been no reason to lug 
in, for example, ivithiUen faile 179, 2909, as a rhyme to cornwayle, 
with scarcely a shadow of excuse from the sense. Or 3) the pas- 
sages containing deled, to deyle, may be corrupt. For this there is 
some ground. The passages are : 

But hwan he haueden }?e kiwing deled, 

And fele sij^es kaueden wosseyled. 1736 

Hwe))er he sitten nou, and wesseylew. 

Or of ani shotshipe to-deyle. 2098 

The first line contains at least one corrupt unintelligible word 
kkving, and not only is the metre of the last Hne unusually defective, 
but the construction to-deyle of for participate in, seems forced and 
unsatisfactory. It would, however, be too hazardous, in the ab- 
sence of parallel passages, to propose any emendation. 
The second passage 

Neuere more he him misdede, 

Ne hond on him with yuele leyde. 994 

cannot be so explained, as dede never appears as deide, and it would 
not be right to conclude that there was an assonance formed by 
calling leyde (leid'e) rather than (laid'e), in face of the older La^a- 
mon forms : laeide, laeiden, leide, laiden, leaide. There was no 
period of English pronunciation in which misdede leyde would have 
rhymed, so far as our researches extend. The passage must there- 
fore be corrupt. In the first place the sense is bad : *' never more 
he hurt him by deed, and never laid hand on him with evil intent," 
merely repeats in the second line what is said in the first. We 

sociate the unusual with the usual old case that e, ey, had the same mean- 
sound, the hesitation which we feel in ing ? At most, they would be diflferent 
selecting one orthography in place of appreciations of the same sound, and 
another, and the variety of pronuncia- might possibly indicate the co-existence 
tions prevalent within a limited dis- of different sounds within the same 
trict, none of which can claim the pre- district. And such coexistence is not 
eminence — true picture of English confined to English dialects. The 
habits of speech in the xiii th century vulgar (een, keen,) coexists with the 
— will make us more readily understand polite (ain, kain) =ein, kein, in Berlin, 
the varieties of orthography adopted Saxony, and many parts of Germany, 
by ancient scribes, and rather admire In the Dyak (Dai-ak) languages of 
than depreciate the partial uniformity Sarawak (Saraa*wak), (^^, ai) constantly 
to which they attained. For myself I interchange even in adjacent house- 
should feel no surprise to find one writer clusters, sometimes even in the same 
representing the " Derbyshire " sound house-cluster, so that (bosee*) or (basai*) 
of sheep, in "ordinary spelling" as would be equally intelligible ioT great, 
sheep, another as shape, and a third as Generally in these languages (ii, ee, ai) 
shipe. Should we then be surprised if interchange on the one hand, and (oo, 
we found an old monk proceeding fi'om uu, an) on the other, as I have just 
a similar district at one time writing been informed (April, 1869) by an 
shep, and at another sheyp ? and should English resident of long standing in 
we conclude in the modern case that Sarawak. See also neither, supra p. 
ee, a, i, had the same sound, or in the 129, n. 1. 

§ 1, No. 5. HAVELOK — XIII TH CENTURY. 475 

want the sense, ''he never more wronged him by word, or deed." 
This is supplied by reading misseyde for misdede, and of the correct- 
ness of this reading we can have no doubt after considering the 
parallel passages. 

Ne found he non that him misseyde, 

!N[e] with iuele on[ne] hand leyde. 49 

Roberd hire ledde, J^at was red, 

pat hau[ed]e J^arned for hire ]?e ded 

Or ani hauede hire misseyd, 

Or hand with iuele onne leyd. 1686 

Me wore leuere i wore lame, 

]?a^ne men dide him ani shame, 

Or tok, or onne handes leyde, 

Vn-ornelfke [vn-ornelike ?], or same seyde. 1938 

The first instance 

Hauelok, J^at was })e eir 

Swanborow, his sister, Helfled, the tother. 410 

is also corrupt on the face of it,^ for the second line of the couplet 
is outrageously prolonged. The word eyr occurs not unfrequently 
at the end of a line, as 110, 288, 605, 1095 and always rhymes 
ynih-fair. This suggests the reading 

Hauelok, that was the eir, 

Swanborow, Helfled her sister fair, ^ 

which at least preserves metre and rhyme, and is immediately sug- 
gested by the parallel passage : 

Of his bodi ne haude he eyr 

Bute a mayden swijje fayr. 110 

The rhyme i, e, as : bidde stede 2548 is frequent. Shewed 
knawed 2057, must be considered in connection with : she we 
lowe 1698, and lowe awe 1291, where lowe, ags. hlaw, means 
a hill, preserved in the Scotch law; as well as with the not 
unfrequent interchange of e, o, as : sore wore = were 236, wore 
= were more 1700, were sore 414, (where Mr. Skeat reads wore), 
more there = there 921, cle[r]k yerk = York 1177, and also of 
0, a: longe gauge 795, 2586, sawe wowe = wall 1962, 2142, 

^ "Corrupt? Lines 410, 411 do not at first proposed: Swanborow, Helfled 

rime well together." Skeat. his sisters fair, in order to preserve as 

2 "We may even imagine how the much of the original as possible, but 

extraordinary error in the MS. arose. the examples : hise children yunge 368, 

Suppose, as usual, that the scribe was we aren boj^e Ipine 619, kniues longe 

writing from dictation. The reader 1769, hundesteyte 1841, wundes swij^e 

gives out: "Swanborow, Helfled her grete 1898, monekes blake 2520, shew 

sister fair," the scribe writes " Swan- that : his sisters faire, would have been 

borow, his sister; " altering her to his required and this would have militated 

as a matter of course, because only a against the rhyme. Unless, indeed, the 

masculine noun had preceded; the reader author could have dispensed with this 

sees the error and exclaims, " Thou hast final e if the necessity of rhyme lay on 

forgotten Helfled thet other ; " the him, as he does dispense apparently 

scribe immediately claps down the words with an e, which is at once plural and 

"Helfled the tother," and is quite dative, in : 

satisfied he has correctly followed the Hwan he hauede mawrede and oth 

reader in the monstrosity: "Swan- Taken of lef and of loth. 2312 

borow his sister, Helfled the tother !" where however perhaps: othe, lefe, 

Se non e vero, e ben trovato, I had lothe, should be read. 


thare = thore = there more 2486, open drcpen = hill 1782. 
We have then to admit that the pronunciation of the writer 
varied in the same word at different times, and that he allowed 
himself* to interchange ^, «, o. The same interchange of {ee, oo) is 
ohservable in the modern Scotch and English : aik oak, aits oats, 
aith oath, caip cope, claith cloth, craik croak, daigh dough, dail dole, 
gaist ghost, gait goat, grain groan, graip grope, hail whole, haim 
home, kaim comb, laid load, laird lord, laith loath, main moan, mair 
more, maist most, raid road, raip rope, saip soap, sair sore, spaik 
spoke of a wheel, taid toad. In Aberdeen we even find (stiin, 
biin) for stoney hone. But it will be seen on examining other 
Scotch ai = {ee) forms, that they often derive from an ags. a, e. 
Herein then we seem to have an indication of the key to this 
dialectic peculiarity. The original (aa) was at one time broadened 
into (oo), and at another squeezed into (ee), and the habits of the 
speaker became so uncertain that all three forms in (ee, aa, oo) 
were in sufficiently common use to allow a rhymester to employ 
whichever was most convenient, till at last (oo, ee) interchanged 
without the intervention of an original (aa). 

We find the regular interchange of «^, ei, as : at hayse = at ease 
preyse 59, deye preye 168, seyl nayl 711, ay domesday 747. 
There seems to be even a probability of seint having been occasion- 
ally dissyllabic, as supra p. 264. Thus, comparing ion 177 : 

In al denemark is wimmaw [non] = (In al Denmark* is wnm-an noon, 

So fayr so sche, bi seint iohan. 1719 Soo fair so shee, bi saa'int Dzhon. 
But gaf hem leue sone anon But gaa* -em lee-ve soon anoon*. 

And bitauhte hem seint Iohan. 2956 And bitaut* -em saa*int Dzhon). 

We have also occasionally the {i) value of u. In two instances 
this value is apparently given to u vo. words which were un- 
doubtedly generally pronounced with (^*), as : 

So \dit ]?ei nouth ne bUune 

Til \2ii to sette bigan J^e sunne. 2670 

per was swilk dreping of ]?e folk 

pat on J^e feld was neu^re a polk 

pat it ne stod of blod so fal, 

pat ])e strem ran iwtil ]?e hul. 2684 

In the first case read so \at \ei \_stunte~] nouth ne hlunne, the 
ags. forms, stunte, hlunne, making metre, rhyme, and construction, 
perfect. In the second, hul, which was supposed by Sir F. 
Madden to mean hill, is perhaps a provincial pronunciation of 
the ags. and old norse hoi, Swedish hoi, Danish hul, a hollow for 
the valley, as the battle was fought at Tetford, near Homcastle. 
But the line is possibly corrupt, and there is no obvious means of 
correction from the want of parallel passages.^ 

1 As it stands the passage must be incKnes to hul hollow, on account of 

translated: "There was such slaying the Scotch use of kowe (hoou, nau), a 

of the people, That on the field there direct descendant of a previous (hwI), 

was never a puddle, That it stood not as opposed to knoll, for a small valley 

so fall of blood, That the stream ran or depression. Part of a village in 

into the hollow(?)." Mr. Murray, who Teviotdak is called Huole-o-the-Burn 

suggested the insertion of stunte above, (hm'1, nuiol, hmeI, Hual). 


The other rhymes do not require particular notice. The value 
of the letters is clearly that established for the xm th century, by 
previous research, with, in the case of om, an anticipation of the 
usages of the xiv th. The metre is rugged and the spelling irregular, 
so that the use of the final -e cannot accurately be determined. 
But there is no reason to suppose it different from what had been 
found for others. 

The orthography of the guttural in connection with t is very 
remarkable, as: knict 239, knicth 77, knith 1068, kniht 2706, 
brouth 336, brihte rithe 2610, bitawte authe 1409, etc., implying 
a peculiarity of pronunciation, which, in the absence of parallel 
usage, and determining rhymes, cannot be appreciated with certainty. 
We must not forget, however, that sigh, droughty height, were 
sometimes called (saith, drAAth, noith) in the xvii th century (p. 
212), and that Keighley in the West Riding of Yorkshire, and 
therefore likely to be somewhat inclined to the same pronunciation 
as the writer of Havelok, is now called (Kiith'l/), and the pro- 
nunciation (nekth) for height, has been noted near Ledbury in 
Herefordshire, which greatly resembles -cth in knicth. At first 
sight -th looks like a metathesis of ht, just as we find ihc 1377 
for ich, and this in connection with the actual occasional oc- 
currence of -ht or even -ct, -cth, would lead directly to the 
usual (-kht) pronunciation. But an examination of the ortho- 
graphy in the poem shews a systematic avoidance of the guttural 
except in relation to t. In all other cases it is expressed only 
by y iy w u, as : eie, fleye, heie, leye = mentire, seyen, sleie ; 
awe = possess, dawes = dags, "drawen drou, fawen =fain, flow, 
galwe, mowe, slou, J?ou = though. Even with t the sign of the 
guttural is frequently omitted, as : ante laute 743, but : awcte 207, 
lauthe 1673. It seems then very possible that these -ct, -cth, -th, -t, 
only mean t, with a merely orthographical indication of the gut- 
tural. This pronunciation of final -cht is not unknown in German.^ 
The otiose h after initial t, and even elsewhere (supra p. 473, 1. 8), 
found occasionally in various manuscripts, but never systematically 
carried out, is not to be compared with this use of h in connection 
with final t, where in most other MSS. the guttural is inserted as 
^> 9^ h^ W^ must also recollect that in MSS., as we have had 
occasion to see also in the Prisoner's Prayer and elsewhere, the 
letter h is used very loosely, even when initial. In Havelok it is 
unnecessarily prefixed in: holde 30, hete 146, het 653, but: 
et 656, heuere 17, her 229, hof 1976, helde 128, etc., etc., and 
we find it omitted in: aueden 163, osed 971, etc., but with no 

1 " Ch lautet gar nicht vor t Ober- 2 The French the, German Thee has 

Rhein und Donau Gebiet, Land und (t) or if it is more dental (.t) on the 

Stadt, (-it, -at) Endsylbe -icht, (-let, continent more than with us, this ap- 

-bt) Endsylbe -licht, (nit net) nicht, plies to every ^ and not merely to those 

ostlech. Rah, Land, (fait-n) Feuchten, written th. In one dialect of the Peak 

Fichte, (Furt) Furcht, (knet) Knecht, of Derbyshire (.t) is heard only, but 

(Hat) Liecht, (Nat) Nacht, (rEat) recht, always, before r and -er. 
(shlBat fed eln) schlecht fechteln, (brat) 
gebracht." Schmeller, Mundarten 
Bayerns, art. 432, 



Chap. V. 

sort of uniformity. Hence the temptation to use it as an idle 
letter, or an orthographical expedient. 

That long i was (ii) or (n) appears among other passages from 

Als she shulde hise clothes handel 

On forto don, and blawe J^er ^ fir ( = Jire) 

She saw therinne a Hth (= light) ful shir (= sheer). 586 

Al so brith, al so shir, 

So it were a blase of fir. 1253 

The word sheer, Gothic skeirs (skiirs) bright, clear, old Saxon 
slciri, middle high German and new low German schir, new high 
German schier (shiir), old high German scieri (skii'ri ?)y, ags. scir 
old norse shir (skiir), Orrmin shir, is a word which from the earKest 
times and in almost all dialects, and specially in English, has re- 
tained the sound of (-iir), and hence is an excellent rhyme to deter- 
mine the old sound of fir. 

The reader will find many points of orthography and pronuncia- 
tion touched on with great care in Mr. Skeat's edition §§ 27 and 28, 
and a full consideration of the treatment of final ^ in § 29.^ 

It is with great diffidence that I annex an example of this difficult 
provincial poem. The text is given exactly, in the pronunciation I 
have ventured on a few alterations, intended to be corrections. 

Havelok 2312-2345. 

Hwan he hauede mawrede 

Taken of lef and of loth. 

Ybbe dubbede him to knith. 
With a swerd ful swife brith. 
And ]7e folk of al \q lond 
Eitauhte him al in his bond, 
J?e cunnriche euml del. 
And made him king heyHke and 

Hwan he was king, Jer mouthe 

mew se 
pe moste ioie j^at mouhte be : 

Conjectured Pronunciation. 
and "Whan ne navde manreed* and 

Taak'en of leev and [ook] of 

TJb'e dub'ed nim to kniit, 
With a swerd ful swidh'e briit, 
And dhe folk of al dhe lond 
Bitaut* -im al in [toe] nis bond 
Dhe kin'eriitsh'e evril deel, 
And maad -im kiq nai-liik and 

Whan nee was kiq, dher mout'e 

men see 
Dhe most'e dzhoi'e dhat mout'e 

bee : 


"When he had homage and oaths 
Taken of dear and [eke] of loath (ones), 
TJbbe dubbed him (to) knight, 
"With a sword ful very bright. 
And the folk of all the land 

^ Mr. Skeat reads ])e. 

2 Mr. Skeat having requested me to 
read and comment on some of these 
points, I endeavoured to do so, in great 
haste, at a time when accidental circum- 
stances disabled me from given them 
proper attention. In those cases where 
the present statements diflFer from those 
hasty expressions of mine which Mr. 
Skeat, anxious not to smother opinions 

Committed to-him al in[to] his hand 
The kingdom every part, 
And made him king, highlike and wel. 
When he was king, there might one see 
The most joy that might be ; 

opposed to his own, has politely printed, 
they must be considered as corrections, 
resulting fi'om careful re-examination. 
I regret not having been able to examine 
all the cases of final ^, to determine 
the circumstances of its elision and 
suppression, but I believe that it was 
not otherwise treated than in the Cuckoo 
Song and Prisoner's Prayer. 

§ 1, No. 5. 




Euttinge with sharpe s-peres, 

Skirming with taleuaces, ]>at 

mew beres, 
Wrastling with laddes, putting 

of ston. 
Harping and piping, ful god won, 

Leyk of mine, of hasard ok, 

Eomanz reding on ]>e bok ; 

per mouthe men here fe gestes 

pe gleymen on ]>e tabour dinge ; 

per mouhte men se fe boles 

And ]7e bores, with hundes teyte ; 

po mouthe men se eueril gleu, 
per mouthe men se hw grim 

greu ; 
"Was neu^re yete ioie more 
In al fis werd, fan fo was ]7ore. 

per was so mike yeft of clones, 

pat fou i swore you grete othes, 

I ne wore nouth j7er-offe croud : 
pat may i ful wel swere, bi god! 
pere was swi]7e gode metes. 
And of wyn, J;at men fer fetes, 
E-ith al so mik and grete plente, 
So it were water of ]7e se. 
pe feste fourti dawes sat. 
So riche was neu^re non so fat. 

But'iq* [dher was] with sharp'e 

Skirm'iq* with tal'vases, dhat 

men beer'es, 
'Rweist'liq' with ladz, put'iq* of 

Harp'iq* and piip'iq*, ful good 

Laik of Miin, of Has'ard ook, 
Hoom'ans* reed'iq* on dhe book ; 
Dher mout*e men see "re dhe 

dzhest'es siq'e, 
Dhe glai'men on dhe taa'bur 

diq'e ; 
Dher mout'e men see fe bol'es 

And the boo 'res, with Hund'es 

tait'e ; 
Dhoo mout'e men see evril gleu, 
Dher mout'e men see huu Grim 

greu ; 
Was never jet'e dzhoi'e moor'e 
In al dhis world, dhan dhoo was 

Dher was so mik*e jeft of 

Dhat dhou i swoor*e ju greet 

In'e woor'e nout dherof'e krod : 
Dhatmai i ful welsweer'e, biGod! 
Dher was swidh'e good'e meet'es, 
And of wiin, that men fer fet'es, 
Hiit al soo mik and gret plen'tee* 
Soo it wer waa'ter of dhe see. 
Dhe fest'e foour'ti dau'es sat, 
So ritsh'e was never noon so 



Butting [there was] with sharp spears, 
Fencing with shields that one bears, 
Wrestling with lads, putting of (the) 

Harping and piping, full good quantity. 
Game of Mine, of Hasard eek, 
Romance reading on the book. 
There might one hear the jests sung. 
The gleemen on the tabour drum, 
There might one see the bulls baited, 
And the boars, with merry [staunch P] 

Then might one see every glee, 

There might one see how Grim grew; 
Was never yet joy more 
In all this world than then was there. 
There was so great gift of clothes 
That though I swore you great oaths 
I- (not) were not thereof oppressed : 
That may I full well swear, by God. 
There were very good meats, 
And of wine, that one far fetches, 
Right also much and great plenty, 
As-if it were water of the sea. 
The feast fourty days lasted, 
So rich was never none as that. 


6. King Hoen, circa a.d. 1290. 

The story of King Horn exists in three several manuscripts which 
present such great varieties both of orthography and language, that 
the text must be considered uncertain. The oldest ^ was apparently 
written about the latter half of the xiuth century, and is that 
which will be followed here. In some cases f occurs for j or 2 
which represents 5. On this orthography see supra (p. 464). The 
dialect is Midland, and the whole poem bears a great affinity to 

There is the usual rhyming of e, e or u, e when u statids for i : 
adrenche ofj^inche 105, Westemesse blisse 157, ire = ear were 
309, wille telle 365, pelle fulle = pall Jill 401 , brunie = armour 
denie = din 591, dunte wente 609, ferde hurede 751, custe = kissed 
reste 1189, etc. 

There are a few cases of e,. <?, in which the a should be replaced 
by e, as : biweste laste 5, warne berne 689. 

As in Havelok, there are cases of e, 0, in which one or the other 
letter must be dialectically altered, if the readings are correct : 
more ^ere 95, swerde orde 623, sende yilonde 1001, posse Wester- 
nesse 1011. We have a, in: felawe knowe 1089. 

A few cases of w, 0, may shew a dialectic pronunciation of u as 
(0), or as {u) : stunde londe 167, ])ojte fujte 277, buje iswoje 
427, pnge isprunge 547, hunde fonde 831. 

In some cases u = (uu) seems to rhyme with u = (yy). In 
bur mesauentur 325, 649, bure couerture 695, one might fancy 
that the French word was mispronounced with (uu). The word 
lure 270, might therefore be to lure, which makes good sense, and 
have been used as a term of falconry, but would then, probably in 
a Saxon's mouth, have been called (luur'e), but it must apparently 
have been to lower or watch for,^ which would be properly (luur'e), 
since the Harl. MS. 2253, fo. 85, reads loure. iStuard 275, 393, 
is probably a clerical error for stiuard compare ags. stiward, which 

1 Cambridge Univ. Lib. Gg. 4, 27, 2. ihc ich y I 

This is contrasted with the Bodleian jou you ou you 

MS. Laud 108 fo. 219^, and Harl. MS. laste sg., lesten pi., yleste sg., last 

2253, in the preface to : King Horn, fairer feyrer feyrorer fairer 

withFragmentsof Floriz andBlaunche- rein reyn reyne rains. 

fleur, and of the Assumption of our Lady, miste micte mihte m ight 

from a MS. (G^. 4, 27, 2) in the Cam- birine upon-reyne by-ryne rain upon 

bridge University Library; also from brijt brict bryht bright 

MSS. in the British Museum. The flur flour flour flower 

Assumption of our Lady (Add. MSS. colur colur colour colour. 
10036) and Fragments of the Floyrea 

and Blancheflur (Cotton Vitellius D. "^ "lure(n), 0. Dm^cA leuren, loren, 

iii), edited with notes and glossary by Fr. leurrer, lure, Chauc. G. t. 5997 ; 

J. Eawson Lumby, M.A. London, lured {part.) vis. P. P. 3351. — (luren) 

1866. 8vo. pp. XX, 142. E. E. T. S. lourin, L. Germ, luren {speculari}) lour 

The extracts from the three MSS. taken {lower) scowl, prompt . parv>. 316 ; loure 

in the above order present the follow- Gow. conf. am. 1, 47 ; Rich. 3470 ; via 

ing among other varieties, P. P. 2735; Triam. 1032; louring 

he he heo they {part.) Chauc. C. t. 6848." Strat- 

beon ben ben be mann, 373. 

§ 1, No. 6. KING HORN — XIII TH CENTURY. 481 

occurs 227, and is the reading of the Harl. MS. 2253 elsewhere. 
In : ture pure = tower peer 1091, we must suppose pure={])u\Lre)j 
to pore or look intently The origin of the word is very obscure. 
The reading of the Harl. MS. 2253 is totally different, and intro- 
duces loke for pure. 

The form ou occasionally, but very rarely occurs, by no means so 
frequently as in Havelok, is: galun glotoun 1123, harpurs gigours 
1471. This applies only to this particular MS. of King Horn. Pro- 
bably the ou is fully as frequent in the Laud. MS. 108, as it is in 
that MS. copy of Havelok, both these poems being in the same hand- 
writing. The greater rarity of ou in this Cam. MS. of King Horn 
is evidence of its greater antiquity, and forms a presumption in 
favour of earlier copies of Havelok having also existed. It is cer- 
tainly desirable for the investigation of the orthography and develop- 
ment of the English language in the xin th century, and especially 
with a view to illustrate Havelok, to have the Laud MS. copy of 
King Horn accurately printed and compared with the Cam. MS. 
The scribes of the two MS. possibly belonged not only to dif- 
ferent times but to different districts, and yet were so nearly con- 
temporary, that the comparison would probably clear up many 
points of difficulty. In the Harl. MS. 2253, ''which has been 
printed, but very badly, by Ritson in the second volume of his 
Metrical Romances," (Lumby, p. vi.) the ou is paramount. 

Sometimes a word is changed for the sake of the rhyme, as ; 
birine = he-rain bischine 11, j\q = ethe = easily dife = dethe = 
death 57, ires = ears tires = tears 959. The two latter are how- 
ever perhaps rather to be considered as dialectic peculiarities. 

^Notwithstanding all these resources the shortness of the lines 
seems to have driven the rhymester to great shifts, unless the scribe 
has much belied him, for we have such decidedly false rhymes as : he 
deie 331, fofte brijte 389, bij^o^te mijte 411, jonge bringe 279, ringe 
^onge 565, 1187, (query, read p'nge, the form found in the Harleian 
MS. 2253,) sede read seide leide 691, heirs read heiris pris 897, 
his (?) palais 1255, yrlonde fondede read fonde 1513, queue beon 
1519. To these we must add: bure foure 1161, unless we admit 
for (fuu're) (foou're) as supra p. 446, 1. 21. It is however pro- 
bable that all these cases are mistakes. The great diversity of the 
MSS., forbids us to lay great store by any particular readings. 

The marked peculiarity of the poem, and one which makes it 
worth while to notice it especially, is the prevalence of assonances, 
single, or double, that is, assonances in which the consonants after 
the identical accented vowel are different, but those, if there are 
any, following the identical unaccented vowel are the same or 
different, as in Spanish ; and assonances which being half rhyme 
and half assonance, may be called conassonances, the accented sylla- 
bles rhyming, and the unaccented being assonant, which also occur 
in Spanish though they are not legitimate. Compare the as- 
sonances of dissyllables and monosyllables in King Alisaunder, 
supra p. 452, note, col. 1, 1. 13. These assonances, which are so 




Chap. V. 

clearly developed in King Horn, remove any difficulty about ad- 
mitting them in Havelok, where they arc not so frequent. The 
following is a list of both kinds. 

Assonances: sones gomes 21, beste werste 27, gripe smite 51, 
admirad bald 89, makede == mak'de uerade 165, swi]?e bliue 471, 
whit ilik 501, proue woje 545, take rape 553, trewe leue 561, 
man cam 787, woje gloue 793, nadde harde 863, rynge Eymen- 
hilde 873, 1287, compaynye hije 879, shorte dorste 927, blife 
bliue 967, iknowe oje 983, haue felaje 995, blowe j7roje 1009, loje 
rowe 1079, wunder tunge 1247, grauel castel 1465, yswoje louje 
read loje 1479. 

Conassonances : moder gode 145, gumes icume 161, doster read 
dojter fo^te 249, scholde woldest 395, lijte knijtes 519, feste 
gestes 521, igolde woldest 643, dojter ofte 697, ride bridel 771, 
ariued fine 807, fijte kni^tes 811, borde wordes 827, hundes funde 
881, knijtes wijte 885, dorter lofte 903, while bigiled 957, knijtes 
fi^te 1213, houe proued 1267, draje fela^es 1289, hundred wunder 

The rhyme : time bime 533, is interesting from its association 
with the same rhymes in Chaucer and Gower (p. 280). 

The word pleinc/ 32, seems to be a contraction oi pleying, and this 
renders the rhyme : king pleying 32, perfect. 

The following may serve as a specimen of the language of this 
poem, according to this more ancient version. The pronunciation 
indicates occasionally conjectural emendations, principally for the 
sake of the metre. 

King Rom 223-234, 241-276. 

J?e kyng com in to halle 

Among his knijtes alle : 

For]? he clupede aJ7elbrus, 

pat was lliward of his hus. 

Stiwarde, tak nu here 

Mi fundlyng for to lere 

Of fine mefter^, 

Of wude and of riuere, 

And tech him to harpe 

Wi]? his nayles fcharpe, 

Biuore me to kerue 

And of \q cupe feme. 

Ailbrus gan lere 

Horn and his yfere : 

Horn in herte la^te 

Al fat He him tajte. 

In fe curt and ute 

And elles al abute, 

Luuede men horn child. 

And meft him louede Eymenhild, 

Conjectured Pronunciation. 

Dhe Kiq kaam in to Hal*e, 
Amoq* His knikht'es al'e : 
Perth He klep'ed Aa'thelbruus, 
Dhat was Stii'ward of His huus. 
Stii'ward" taak nuu neer'e 
Mi fund'liq, for to leer*e 
Of dhiin'e mesteer'e. 
Of wuud and of riveer*e, 
And teetsh Him to Harp*e 
"With His nail"es sharp "e, 
Bifoor'e mee to kerve, 
And of dhe kup'e serve. 
Aa'thelbruus gan lee 're 
Horn and nis ifee're : 
Horn in nert'e lakht'e 
Al dhat Hee Him takht'e. 
In dhe kuurt and uut'e 
And el'es al abuut'e 
Luvde men Horn Tshild. 
Meest luvd- im Eiim*enhild 

§ 1, No. 6. 



pe kynges o^ene dofter, 
He was meft in J7o^te, 
Heo louede so horn child 
pat ne^ heo gan wexe wild : 
For heo ne mi^te at horde 
Wi]> him fpeke ne worde, 
!N"e noft in J^e halle 
Amowg ])e kni^tes alle, 
'Ne nowhar in non o]jere ftede : 
Of folk heo hadde drede : 
Bi dale ne hi ni^te 
Wif him fpeke ne mijte. 
Hire foreje ne hire pine 
Ne mijte neure fine. 
In heorte heo hadde wo. 
And fu^ hire hifojte fo. 
Heo fende hire fonde 
Afelhrus to honde 
pat he come hire to, 
And eiifo fcholde horn do 
Al in to bure, 
For heo ga.n to lure. 
And ])e fonde feide 
pat fik lai ])at maide 
And had him come fwif e, 
Por heo nas noj^ing hlijje. 
pe ftuard was in herte wo, 
Per he nnfte what to do. 

Hhe kiq*es oogh*ne dokht'er. 
Hir was -e meest in thokht'e. 
Heo luvde soo Horn Tshild 
Dhat Heo gan weks'e wild. 
Por Heo ne mikht at hoord'e 
With Him speek'e noo word'e 
Nee nokht in dhe nal'e 
Amoq* dhe knikht'es al'e, 
l^ee in noon oodh're steed'e. 
Of folk Heo Had'e dreed'e. 
Bi dai"e nee bi nikht'e 
With Him speek neo ne mikhte. 
Hir sor'ghe nee nir piin'e 
Ne mikht'e nevre fiin*e. 
In Hert Heo had'e woo. 
Dhus nir bithokht-e dhoo, 
Heo fende Hire sond'e 
Aa'thelbruus to Hond*e, 
Dhat he kuum nir too. 
And al'so shold Horn doo 
Al in too nir buu*re, 
For Heo gan to luu*re. 
And dhe sond*e said'e 
Dhat sik lai dhat maid'e 
And bad him kuum*e swiidh'e^ 
For Heo n-as noo-thiq bliidh'e. 
Dhe Stii'ward was dher woo, 
For He nust'e what to doo. 


The king came in to hall 
Among his knights all. 
Forth he called Athelbrus 
That was steward of his house. 
" Steward take now here 
My foundling, for to teach 
Of thy ci'aft, 
Of wood and of river, 
And teach him to harp' 
"With his sharp nails, 
Before me to carve, 
And serve of the cup." 
Athelbrus began to teach 
Horn and his companions-. 
Horn received in his heart 
All that he taught him. 
In the court and out 
And else all ahout 
Loved one Horn Child. 
Most loved him UimenhUd, 
The king's own daughter. 
To-her was he most in thought. 
She loved so Horn Child 
That she began to grow wild. 

For she might not at table 
With him speak no word, 
Nor nought in the hall 
Among all the knights. 
Nor in no other place. 
Of people she had dread. 
By day nor by night 
With him she might not speak. 
Her sorrow nor her pain 
Might not ever cease. 
In heart she had woe. 
Thus bethought her then. 
She would-send hir messenger 
To the hand of Athelbrus, 
That he should come-to her, 
And thus should bring Horn 
All into her bower. 
For she began to lower (lure ?) 
And the messenger said, 
That sick lay the maid 
And bad him come quickly (?) 
For she was in no wise blithe. 
To-the steward was woe, 
For he knew-not what to do. 


7. Moral Odb, Pater Noster, Orison, end of xiith Century. 

The compositions of the xrrr th century have all a decidedly local 
character, but the phonetic meaning of the letters, which is all we 
have to deal with, seems as firmly established as in the xivth. 
The poems mentioned above belong perhaps to the xn th century. 
The copies to which we shall refer have been published for the 
Early English Text Society.^ It will not be necessary to examine 
them in much detail. They present much the same character as 
Havelok, with the e, i and e, o and o, a rhymes. The orthography 
is very unsteady, and it is difficult to feel certain in any place that 
we are not dealing with a scribal error rather than a peculiarity of 
pronunciation. It will be sufficient to deal with a few peculiarities. 

The Moeal Ode, or Poema Morale : Rowen sowen = rue sow 
19, written : ruwen seowen, in the Egerton MS., are ags. hreowan, 
sawan, and can only rhyme by the dialectic interchange of e, o, as : 
shewe lowe, in Havelok (supra p. 476). Seide misdede 129, seiden 
reden 223, require a peculiar dialectic pronunciation of seide as sede, 
and that this existed we learn not only from the orthography : of 
sede, rede 155, in this MS. but from the parallel rhymes : sede 
misdede 131, sede rede 225 in the Egerton MS. See supra, p. 447. 
Hulde felde 343, hulle fulle 347 and durlinges 385, are examples 
of the use of u for «, or «, common in this MS. 

The Pater Noster offers many examples of ii for i: vrale 14, 
of-funche^ 16, ufele 17, fenne wunne = win 19, inne sunne = 
sin 23, 139, 224, wulle ifulle 55, sunne unwune 282. The rhyme : 
bone clene 167, shews how o was written for e even when e was 
pronounced. Wrei^ segge^ 179, shews the derivation of the (ai) 
sound from (e^h), and : mei dei 169, shews the identity of ei, ai. 

The Orison, or On God Ureison oe Ure Lefdi, contains a few 
peculiarities which suggest scribal errors : Marie lefde 1, lefdi liuie 
11, lefdi beien 17, could not have rhymed. The first would be 
satisfied by the more ancient form lefdie, ags. hlaefdie, which is 
justified by lafdie in Layamon, 15647, or else by the contracted 
form Mariy which we have already had reason to suspect, p. 441. 
The difficulty of: lefdie beie 17, as it would then be written, is 
the same as that of : beie offrie 2, and : lefdie liuie 1 1 offers a 
singular form for Hue, and a transmuted accent. See several other 
instances of like forms, supra p. 446. See also the infinitives in 
the Assumpcioun in Lumby's King Horn, p. 44, and in Dan MicheFs 
Ayenlite. Kwene reine = queen rain, 57, should evidently be : 
kwene rene, the old ags. form ren, which existed as well as regen, 
here coming into use. 

1 Old English Homilies and Homi- with introduction, translation and notes 

letic Treatises (Sawles Warde, and >e by Eichard Morris, 1867-8. The Mo- 

Wohunge of Ure Lauerd : Ureisuns of ral Ode is No. 17, p. 158, and a dupli- 

TJre Louerd and of Ure Lefdi, etc.) of cate of the first 270 lines from the 

the Twelfth and Thirteenth Centuries, Egerton MS. is given in an Appendix, 

edited from MSS. in the British Mu- p. 288. The Pater Noster is on p. 55, 

seum, Lambeth and Bodleian Libraries, and the Orison on p. 191. 

§ 1, No. 7. 



The following brief extract from the Paternoster will convey 
some notion of the language. 

Paternoster^ 75-98. 

Adueniat regnum tunm. 

Cume ]>i riche we segge'^ hit. 

Hercni^ alle to J^is writ, 
his riche is al fis middeleard. 
Eor^e and heofene and uwilcherd 
ofer alle is his muchele mihte. 
lauerd he is icleped mid rihte. 
Lauerd he is of alle scafte. 
In eor^e. in heuene is his mahte 
alle fe scafte ]>e he bi-gon. 
fet is f et sod^e hit wes for mon 

alle J^iwge he makede set agan. 

Er he efre makede mon. 

he makede mon i rihtwisnesse. 

Onlete on his onlichnesse, 

Alle dor and fujel ifliht :f 

lete he makede adunriht. 

fene Mow he lufede and welbi- 

and for-J^i his neb upward he 

]>et wes al mid muchele skile ^ 
^if he hit understondon wile. 
Neb upwardes he him wrohte. 
he walde ]>et he of him ]7oht[e]. 

Al swa ]7e lauerd ]>et him wrohte. 

Conjectured Pronunciation. 
Adveeii'iat re g* num. tuu'um. 

Kuum'e dhi riitsh'e ! We sai*eth 

Herk'nith al*e too dhis rwit. 
His riitsh is al dhis mid'el erd, 
Erth and nevn- and ii'wilk nerd. 
Over al is His mutsh'le mikht'e 
Laverd ue is iklep'ed mid rikht'e 
Laverd ne is of al'e skaft'e. 
In erth, in heven is His makht'e : 
Al-e dhe skaft'e dhee He bigon*, 
Dhet is dhet soodh, nit wes for 

Al*e thiq He maaked [? ?] 
Eer He evre maak'de mon. 
He maak'de mon i rikht'wisnes'e, 
On'leet on His on*litshnes*e. 
Al'e door and fuugh*el iflikht* 
Leet -e maak'ed aduun*rikht : 
Dheen-e Mon He luvd- and wel 

And fordhii* His neb up "ward* he 

Dhet was al mid mutsh'le skiile, 
Jif je Hit un'derstond'on wiil'e. 
Nebup'ward'es nee nim rw;okht*e. 
He wald'e dhet nee of nim 

Alswaa dhe Laverd dhet nim 


Mr. Morris's Translation. 

Adveniat regnum tuum. 

Thy kingdom, come, we do say it, 

Hearken all unto this writ ! 

His kingdom is this middle earth, 

Earth and heaven, and each abode ; 

Over all is his great might. 

Lord he is called with right ; 

Lord he is of all creatures, 

In earth and heaven is his might. 

All the creatures that he formed. 

That is truth, it was for man, 

All things he made to appear 

Before he ever made man. 

He made man in righteousness, 

In the form of his own likeness. 
All deer (animals) and fowl of flight 
He made to stoop adownright (down- 
Man he loved and cared for well. 
And therefore his face upward he 

That was all for a good skUl (reason), 
If that understand ye will. 
Face upwards he him wrought. 
He would that man of him thought. 
That he should love him with thought 

(in his mind) 
As the Lord that him wrought. 


^ 2. Unrhymed Poems of the Thirteenth Century and Earlier. 

The rhymed poems having resulted in a satisfactory deter- 
mination of the values of the letters, it is necessary to apply 
the result to the examination of documents in which no 
rhyme is employed. The first of these that has been selected 
is so careful in its orthography that it is in many respects 
more fitted for our purpose than the laxly written poems 
already considered. The second has chiefly antiquity to 
recommend it, and its principal phonetic value lies in the 
great diversity of representations which it supplies for the 
same word. 


Orrmia's Orrmulum^ is written in a strict orthography, with 
some inevitable sHps here and there perhaps, which escaped the 
author's evidently careful and repeated revision,^ and as the object 
of this orthography was phonetic, the poem may be fairly considered 
as beirig the first example of the application of the purely phonetic 
principle in the orthography of English. 

Orrmin's scheme was to double the following consonant when a 
vowel was short. The origiQ of the feeling which led to this no- 
tation has been already explained (p. bb). This plan has the ob- 
vious disadvantage of not indicating the length of a vowel when no 

^ The Ormulum. Now, first edited italic, we have in these sixteen lines, 

from the original manuscript in the broj^err (twice), trowi^j^e, takenw, 

Bodleian (Jun. MS. 1.) with Notes and rejAell, follj/ienn, swasumw (twice), 

a Grlossary by Robert Meadows White, biwwille, wenwd, litde, hafe]?^. As we 

D.D, Oxford, 1852, 2 vols. 8vo. "If nave also at length broj^err (twice), 

we consider alone the character of the Wallt', afi'f, flseshess, crisstenndom, 

handwriting, the ink, and the material ])urrh (three times), fulluhht, godess, 

used by the scribe, we find reasons for jjatt, witt, hafenn, etc., and as in the 

placing the date of the MS. early in cases of superposition the writing was 

the thirteenth century," pref. Ixxii. crowded, I conceive these to have 

Mr. Garnett considers it to have been been corrections, similar to the little ac- 

written in Peterborough. Dr. White cent marks by which words were sepa- 

writes '■^ The Ormulum" with a pre- rated that had been too closely written, 

fixed the and single r in the above If then in some cases we find a single 

title, but in the introduction we read — consonant where we should have ex- 

jjiss hoc iss nemmnedd Orrmulum pected a double consonant, we may 

forr)»i jjatt Orrm itt wrohhte fairly attribute it to a slip which has 

where Orrm is a contraction for Orr- escaped correction. Occasionally, where 

min as we see by the example given two consonants follow the vowel, the 

beloWip. 491 dedication 324. first consonant seems not to have been 

2 In the facsimile of the sixteen doubled, either through the author's 

opening lines prefixed to White's edi- inadvertence or from his not having 

tion, we see that the second consonant thoroughly settled the system of writ- 

in a reduplication was sometimes ing, so that we find kinde smdijinndenn, 

written over the other, and sometimes which must have both had a short t, 

not. The same was the case occasion- and may be compared to the double 

ally with h in ;5h, etc. Thus, repre- forms amang, amanng, which must 

senting the superior consonant by an have signified the same sound. 

6 2, No. 1. ORRMIN XII TH CENTURY. 487 

consonant followed. Thus in the opening lines ]>e, i, o, to, swa were 
all probably short, and ba = both, was long. The writing, how- 
ever, shews no difference. There was also this inconvenience that 
as the short vowels are more frequent than the long, the writing 
was overladen with doubled letters. The expedient of doubling the 
vowel to indicate length, also very common and natural, overcomes 
both difficulties, as may be seen by the example of pronunciation in 
palseotype below p. 490. Thorpe in the Preface to his Analecta 
Anglo- Saxonica, 1846, p. xi, attributes to Orrmin the precise cor- 
respondence of long and short vowels which exist at the present 
day,^ so that according to him Orrmin's a, e, i, o, u represented {ee 
86, ii e, ai i, oo o, uu a), an hypothesis which our preceding inves- 
tigations render untenable. If any weight is to be attributed to 
our determination of the values of a, e, i, o in Chaucer, and u in the 
Cuchoo Song, we can hardly conceive the pairing of the vowels to 
have been otherwise then (aa a, ee e, ii i, oo o, uu u), except that 
very possibly {aa a, ee e, ii i ) may have replaced the first three 
pairs, and as to the last pair, there might, from previous examples, 
be a suspicion that the long and short u may have been at least 
occasionally (yy, y) ; but no examples of the use of u for i, e seem to 
occur, so that u should probably be always read as (uu, u). The 
form ou for (uu) never occurs. 

There are very few divided vowels, but we meet with (B and eo. 
The ce in numerous instances replaces an ags. ea as in : daed dead 
dead, drsem dream sound, rsem hream cvj, taem tea)n offspring, flaerd 
fleard mockery, staep steap steep. It often alternates vtdth e and 
sometimes even with eo, thus we have : draedenn dredenn, 2 pr. 
drsedesst, 3 pr. dredej^f, 2 pi. draedenn, 3 p. di-edde, imp. dred ; 
draefedd, dreofedd, drefedd. These confusions seem to indicate that 
(B, eo, e had the same sound. Even if ce retained its true ags. sound, 
which was probably (aeae, ae), this would readily be confounded with 
(ee, e), and this again with (ee, e). It seems preferable then to give ce. 
the same sound as e, viz. (ee, e), or else to regard ce as (e), and e as (e). 

As respects eo, Mr. White observes that : *' a remarkable instance 
of the preference of e for eo will be found by the omission, nearly 

^ He says : " The author seems to nounced God, not Gode), etc. Thus 

have been a critic in his mother-tongue ; hus is to be pronounced hoos, whereas 

and to [through ?] his idea of doubling ])uss, with a double s, is our thus.'' 

the consonant after a short vowel (as Tyrwhitt, in his Essay on the Language 

in German), we are enabled to form and Versification of Chaucer, Part III. 

some tolerably accurate notions as to § iv. note 52, declares himself unable 

the pronunciation of our forefathers. to comprehend the meaning of those 

Thus he writes min with a single n doubled consonants, and in quoting the 

only because the i is long or diphthonal, commencement of the Dedication, "ven- 

as in our mine. So also in kinde (pro- tures (first begging Ormin's pardon for 

nounced as our kind,) dom, hoc, had, disregarding his injunction) to leave 

lif (pronounced as our life), etc. On out the superfluous letters." To have 

the other hand, wherever the consonant been consistent, then, he should have 

is doubled, the vowel preceding is written : beging, lev, leters, instead of 

short and sharp, as in 7;ett (pronounced the " superfluously lettered" begging, 

as our yet, not yate, as it would be if le«v^, letters ! 
written with a single t) Godd (pro- 


uniform, of o in the latter part of the MS., in the inserted leaves, 
and in the dedication and preface, as in the forms lede, fede, 
werelld, etc., the o having been written in the above words and 
in others in the first part of the MS., afterwards erased, and 
then re-written. In these last named instances the o has been 
retained in printing in order to preserve the orthography. Perhaps 
the was rejected as not essential for pronunciation; Cf. our word 
peopled Of course such deletions and restitutions of o could not 
have taken place unless eo formed one syllable, as White observes, 
quoting v. 8571 : 

\2i shulenn beon off heore kinn. 
Possibly the writing may have been Orrmin's, the deletion his 
brother's, who was requested to examine the manuscript, ded. v. 65 : 

Annd te bitseche ice off j^iss boc 

heh wikenn^ alls itt seme]?]? 
all to ])urrlisekenii illc an ferra 

annd to ]?urrlilokenn offte, 

certainly rather for the purpose of detecting trips in doctrine, 

]jatt upponn all ]?is boc ne be 

nan word jsen Cristess lare, 
nan word tatt swi]?e wel ne be 

to trowwenn annd to foil jhenn ; 

but we can easily imagine *'broferr Wallterr" having extended his 
observations to the spelling, and Orrmin having on further reflection, 
restored his own orthography. In this case Orrmin attached a 
value to eo different from (ee). However it be, we find as a matter 
of fact that in White's glossary almost every word spelled with eo 
has a secondary form spelled with simple e. This would rather 
indicate (ee|_o), with a strongly marked (ee) and an evanescent (o), 
comparable to the (oo[_u, o6*w) in our modem pronunciation of hiow 
= (noou). 

The fonns at, ei, au, ou do not occur, but the syllables ijg, egg, 
a55, aww, eww, most probably indicated the presence of diphthongs. 
The letter 5 had of course a different sound from g. The regular 
(gh) sound seems to have been written 5 A, while (kh) was h or hh. 
Thus from a-^hemi to own, we have ah owns, and ahhte goods, cattle. 
We have also herr'^henn to save, herrhless salvation. Observe that 
in these cases 5A comes before a vowel, as in hall-^he, re'^hell^ 
fol^hmn^ etc., and ^, hh^ before a consonant or at the end of a word, 
and this rule appears to have been consistently carried out. The 
simple 5 then probably functioned as (j), as in : garrken, gate, ge, 
jelden, gellpenn, georne jeorrne geme jerrne, ger, gife, giff, jilt, 
50CC, 50I, sung, jure. The initial gh is peculiar to the word 5A0 = 
she and the contraction ■^h6t= ^ho itt. In the later text of Laja- 
mon we have -^eo for she ; see also ghe, ge, supra p. 467. It would 
be difficult to pronounce 3A0 otherwise than (^ho, jho), and it 
would seem to be a peculiar derivative from heo, the (jh) being 
generated in the same way that it is in a not unusual modem pro- 

^ White translates, office, duty, attendants, and Stratmann sub voce 
charge. See Lajamon's wikenares= wxken. 

§ 2, No. 1. 




nunciation of tlie words, hue, Hume, Hughes = (jIiuti, Jhuum, 
Jhuuz). From these (jho, jhe) forms the subsequent (shoo, 
shee, shii) easily follow. What then was the effect of 5 when 
final ? We know that many orthoepists, as Wallis, consider 
that the final element in the diphthongs (ai, au) is (j, w) and not 
(i, u), p. 186. We see also from the example oi Awwstin, Ded. v. 10, 
which we know from Latin sources must have been (Austiin*), that 
Orrmin belonged to this class. It follows therefore that eww must 
must have been (eu) in cnewwe and that a-gj^, ^jj must have been 
(ai, ei), or (aai, eei), as it is unlikely that Orrmin would have made 
the difference, the duplication of j serving only to shew the strict 
diphthongation of the elements. 

The legitimacy of this interpretation will be more readily 
admitted after an inspection of the following lists of all simple 
words which I have observed in Orrmin containing aj j and ejj. 

da7;3 day, gen. and pi. 

dajhess, dajjess ; ags. 

fa33err/««V, ags. faegr 
faj^re fairly^ ags. fajgere 
frajjnen to ask, ags. freg- 

nan, Lancashire frayne. 
ma^3 (1) may, ags. mseg ; 

(2) maid icel. mey. 
majjdenn maiden, ags. 

majjstre magister 
m&il])Q tribe, ags. maeg'S 
na33 nay 
najjlenn to nail, ags. 

waj3 woe 

waj^n wain, ags. wsegn 
wajjne)?]? carrieth, ags. 


bejjen gen. o/ba both 
bejjsanns bezants 
bej^sc bitter, icel. beiskr 
be^jtenn to beat, ags. 

clsenlej^c chastity 
ezze fear, ags. eg 
ejjlenn to ail, ags. eglan 
ejjj^err either, ags. segj^er 
ej^whaer everywhere, ags. 

fle^^l Jlail, old Fr. flaial, 

Lat. flagellum 
gej^nen to gain, icel. at 

gej^nhke conveniently, 

icel. gegnilega 
idellejjc idleness 
lej^est le33e» lej^^de 

le^^ layest layeth laid 

lej^kenn to play, icel. at 

lej^tenn to inquire, icel. 

at leita 
metlejjc humility 
rejjn rain, ags. ren, regn 
rejjnenn to rain 
rejjsenn to raise, icel. at 

reisa to travel 
sejjst se3j> se^^de 

say est saith said from 

twe^jen twain 

\QZZ i^^fy 

j^ejjm them 
J^ejjre their 
wejje way, age. weg 

lay, from leggenn to lay. 

In almost all these cases we see ajg answering to ags. ag ceg eg^ 
and ej J to ags. eg and once ea, or Icel. ei, and twice e = (je). The 
most remarkable exception is \e%-^m from ags. \am, as it accounts 
for the form \eim, ]iaim, (p. 442, Pater, v. 8), and perhaps for fm, 
forms sometimes found in old English. It does not seem possible to 
establish the transition of ag into ai (agh, Oigh., aj, ai) more clearly. 

The combinations i-^ or «jj occur in -li-^, as innwarrdli"^, 
witerrli^, and in twi^^ess and similar words, where the difference of 
the single j and double jj has to be noted. Properly the sound 
should be that of the very common German termination -ig, as 
inwendig, wahrhaftig, which is theoretically (-i^h) and practically 
(-iX'h), as (in*bhend:i^h, bhaar*Haft:i^A), or (in'bhend:iM, bhaar*- 
Haft:i/;h). It would therefore be hazardous to read ej, «jj, other- 
wise than as (iiX'h, ihh) final or (ii^h, i^h) before vowels. The 
objection that these sounds when final should have been written -/A, 
-ihh, must be met by the habit of the ags. final -ig. The same 
reason may have led Orrmin to use jj in the middle of a word in 


place of J J A, whicli would have been the regular reduplication of 5 A, 
compare ssh in Englissh, dedication 109. The value of uw in -^uio 
is doubtful, but it does not seem likely to have differed from (uu). 
The / between two vowels, and frequently elsewhere, was most 
probably {v), a letter which Orrmin avoids, but ^was of course (f). 
This would accord with the modern AYelsh usage. 

As to the final <?, the rule of pronunciation given, by the strict 
observation of the number of syllables in each line, is precisely that 
at which we arrived for Chaucer, down to the occasional elision of 
an inflectional final e, even when not preceding a vowel, in which 
case Orrmin simply left it out.^ The elisions, however, are not so 
frequent as in Chaucer. Thus, in the first 1000 lines of the Homilies 
in White's text, final e is elided five times before himm, thi-ee times 
before he, twice before himm and Mss, once before hu and once before 
Serodess v. 277, which is very peculiar. The elisions before a 
vowel are more common. Open e perhaps does not occur, so that 
the practice of the end of the xrvth century is justified by an 
English practice at the beginning of the xin th, which, cannot have 
been influenced by l^orman habits. Coalescent words also occur as 
\alde, namm = ]>e aide, ne amm, het = he itt, noff == ne off, nafe, 
naffde = ne hafe, ne haffde, etc. A final d or t changes the follow- 
ing \ to ^, a practice which we have met with before (p. 444, n. 2), 
and which was still preserved in Chaucer's : wiltow = wilt thou, 
etc. (p. 371), but here carried much further. "We may therefore 
feel considerable confidence in pronouncing Orrmulum as follows : 

Orrmuhim, Dedication. Conjectured Pronunciation. 

Annd whase wilenn shall J^iss And whaa'see wii'len shal this 

boc book 

efft ojerr sij^e writenn, 96 eft oo'dher sii'dhe r^6?ii*ten, 

himm bidde ice Jatt het write sim bid ik dhat nee-t rt^ii'te 

rihht ri^ht 

swa summ J^iss boc himm taech- swaa sum dhis book him tEEtsh* 

e}]>f . eth, 

all ]?werrt u't affterr J^att itt iss al thwert uut aft'er dhat it is 

uppo J^iss firrste bisne, 100 upoo* dhis first "e biis'ne, 

wi]?}' all swiUc rime alls her iss with al swilk riim als heer is 

sett, set 
Verbal Translation. 

And whoso shall desire this book All throughout after (the way) that it is 

Again another time to write, 96 On this first example, 100 

Him beg I that he it write rightly "With all such number as is here set 

Just as this book him teacheth, (forth,) 

1 White cites the examples: fra att inne 12739 ; whseroff' 13694, 

mann' to manne 11219 ; to king' 8449, whserofi^e 13704; ofi'wite^hunng 14416, 

to kinge 8370; to grund' 11773, to off wite^hunnge 14617, where I have 

grunde 12547 ; faderr hallf 2269, introduced an apostrophe to mark the 

o faderr hallfe 2028 ; i Godess hus' elision. This omission of e in writing 

625, inn huse 2112; off slap' 1903, sometimes takes place before a vowel, 

off slsepe 3143; j^attlseredd' folic 15876, where it was not necessary according 

jjatt Iterede folic 7440 ; att inn' 12926, to Orminn's system of writing. 

§ 2, No. 1. 




wij^f all se fele wordess ; 
annd tatt lie loke wel ]:-att lie 

an bocstaff write twijjess 104 
ejgwhser J^ser itt uppo J^iss boo 

iss writenn o J^att wise ; 
loke he well J^att he't write swa, 


forr be ne maj j nohht elless 
onn Enngiissh writenn ribbt te 

fatt wite he wel to sofe. 
Annd jifF mann wile witenn whi 

ice hafe don j^iss dede, 112 
whi ice till Enngiissh hafe wennd 

goddspelless halljhe lare ; 
ice hafe itt don forr]?i j^att all 

crisstene follkess berrhless 116 
iss lang uppo fatt an, J^att tejj 

goddspelless hall j he lare 
wij?]? fulle mahhte folljhe rihht, 

]7urrh ]7ohht, ]7urrh word, JTurrh 


* * * :^ 

Ice fatt tiss Enngiissh hafe sett 

Ennglisshe menn to lare, 322 
ic wass, fser J^ser I crisstnedd 

Orrmin bi name nemmnedd. 
annd ice Orrmin full innwarrdlij, 

wiff mu]? annd ec wiyp 
herrte, 326 

her bidde J?a Crisstene menn 

J7att herenn oferr redenn 
fiss boc, hemm bidde ice her Jjatt 

forr me ]?iss bede biddenn : 330 


"With all so many words, 
And that he look well, that he 

One letter write twice, 104 

Everywhere where it upon this book 

Is written on that wise ; 
Look he well that he it write so. 

For he may not else 108 

In English write rightly the word. 

That know he well to sooth. 
And if one will know why 

I have done this deed, 112 

"Why I into English have turned 

Gospel's holy lore ; 
I have done it because that all 

Christian people's salvation 116 

with al see fee*le word'es ; 
and tat ho look'e wel dhat nee 

aan book'staf r2<;ii'te twi^h'es 
ei'whEEr dhEEr it upoo* dhis book 

is Twit'en oo dhat wii'se ; 
look see wel dhat nee-t r^^^ii'te 

forr nee ne mai nokht el*es 
on Eq'lish n^^ii'ten rii?;ht te 

dhat wiit He wel to sooth 'e 
And jif man wiil'e wit -en whii 

ik Haave doon dhis deed'e, 
whii ik til Eq'lish naave wennd 

god'spel'es nal^h'e laa're ; 
ik Haav it doon fordhii* dhat al 

cristee'ne folk*es berkh'les 
is laq upoo* dhat aan, dhat tei 

god'spel'es nal^h'e laa're 
with ful'c makht'e fol^j/h'e riX'ht, 

thurX'h thokht, thur/th woord, 

thur^-h dee'de. 
* :^- *- * 

Ik dhat tis Eq'lish Haave set 

Eq'lish'e men to laa're, 
ik was, dhEEr dhEEr i krist'ned 
Ormiin' bi naam*e nemm'ned. 
And ik Ormiin- ful in'wardli^h 
with muuth and eek with 
Heer bid*e dhaa kristee'ne men 
dhat Hee'ren oo'dher ree'den 
dhis book, nem bid ik neer dhat 
for mee dhis bee'de bid'en : 


Is along of that one (thing), that they 

Gospel's holy lore 
"With full power follow rightly, 

By thought, by word, by deed. * * * 
I that this English have set (forth) 321 

Englishmen to teach, 
I was there where I christened was, 

Orrmin by name named ; 
And I Orrmin full inwardly, 

"With mouth and eke with heart 326 
Here pray the Christian men 

That hear or read 
This book, them pray I here that they 

For me this prayer pray : 330 



Chap. V. 

fatt bro]7err Jjatt tiss Ennglissh dhat broo'dher dhat tis Eq-lish 

writt n^it 

allrairesst wra t annd wrohhte, alrsE -rest Twaat annd rw^okht "e, 

}att broj^err, forr hiss swinnc to dhat broo'dher, for nis swiqk to 

laen, Ieeii 

so]? blisse mote findenn. 334 sooth blis'e moo'te find'en. 

Verbal Translation. 
That brother that this English writing That brother for his labour to reward, 
First of all (men) wrote and wrought, True bliss may (he) find. 

As considerable doubt attaches to the length of the vowel in old 
English, and as Orrmin's orthography is meant to resolve that 
doubt, it seems worth while to collect together all the instances 
where he seems to mark vowels as long. In the following lists, 
which have been collected from White's glossary, all the simple 
(uncompounded) words in which a long vowel before a consonant 
appeared to be indicated with tolerable certainty have been col- 
lected. To all cases in which a vowel is followed by more than 
one consonant, and the first of those consonants is not doubled, 
doubt attaches, because Orrmin's usage fluctuates in some of them, 
and he seems to have thought that two consonants would act oc- 
casionally as well as a doubled consonant. Such words are there- 
fore excluded, as are also all monosyllables ending in a vowel, and 
therefore of undetermined quantity. The use of the short sign (") 
sometimes seems to indicate a short vowel, where only one con- 
sonant follows, and hence a few of the following words may be 
doubtful, but on the whole it would seem that a long vowel was 
intended in each of the following cases. 

List of Oeemin's "Woeds containtn^g Long Yowels.^ 




long A (aa) 






















a J) ess 






























































ran touched 



wat knew 





Long JE (ee) 

This list and the following have been checked by Mr. Brock, 

§ 2, No. 1. 

























































Lonff E (ee) 


















































1 efenn 








































































































Long I (ee) 

























Long (oo) 


































































Chap. V. 




Long U (uu) 









































As considerable interest attaches to the determination of such 
adjectives and substantives as had a final e in early English, and as 
Orrmin's versification establishes with certainty the pronunciation 
of such letters, except when they are elidably situate, I have 
collected from White's glossary all such words, adding the meaning. 
A few substantives are only found in oblique cases, and these are 
marked f because the e may be only inflexional. In the case of 
the adjectives it is not always certain, from a simple inspection 
of the glossary, whether the ^ is a mere mark of the plural or 
of the definite inflection. When I have detected either of these 
to be the case I have omitted the adjective from the list, but I have 
not thought it necessary to verify every case. Such a table of 
German nouns and adjectives would seem ridiculous to a German, 
because he cannot dissociate the e from the words. We have be- 
come so used to its absence that every kind of artificial means is 
necessary to restore the association. 

List of Oeemhst's Adj:ectives aistd Substaittives EKorPTG rsr E. 

adle disease 
8eb8ere clear 
segedef luxury 
sere ear 
sete food 
abbte goods 
ane alone (?adv.) 
ange sorrow 
anndsgete odious 
anndsware answer 
are grace 
arrke arJc 
asse ass 
axe axe 
a^be awe 
bsere bier 
bede prayer 
belle bell 
bene prayer 
benncbef bench 
berrmef barm 
bermef barn 
bettre better 
bilenge belonging to 
birde lineage 
bisne example 
bite morsel 

blisse bliss 

bli)7e blithe 

blome blame 

blostme blossom 

bode command 

bone boon 

bote remedy 

bo)?e booth 

braj^j^e anger 

breme furious 

bridale bridal 

bridgume bride- 

bulaxe axe 

bule bull 

buret bower 

care care 

cbele cold 

cbepinngboj^e mar- 

cbesstre city 

clakef accusation 

clene clean 

cribbe crib 

ciide cud 

cuUfre dove 

cweme agreeable 

daedbote repentance 
dale part 
daerne secret 
dafFte humble 
dale valley 
dede deed 
deme f judge 
deope, depe deep 
deore, dere dear 
dri^je dry 
drubbjiet drought 
dure door 
dwillde error 
ecbe eternal 
Q^^Q t edge 
ebbte eight 
elde t age 
ele oil 

ende end country 
eorj^e, er)>e earth 
errt'e animal 
errnde errand 
ejbe eye 

ejbesallfe eye-salve 
ejbesibbj^e eyesight 
ej^e t fear 
fsewe few 

fallse false 
fasste fast s. 
fele, fele, fele many 
feor]7e fourth 
fere fere power 
fifte ^fifth 
fiftende t fifteenth 
fode food 
forrme first 
frame profit 
fremmde strange 
frofre f comfort 
frummjje beginning 
fulre foul-er 
galle gall 
gate way 
genge gang 
gillte t tribute 
grene green 
gre^^fe herald 
grimme grim 
bsele health 
hsesef command 
baete f heat 
belle hell 
bellfe handle 
bellpe help 

§ 2, No. 1. 



heoffne heaven 

heore their (pron.) 

heorrte heart 

here host 

hete, liete hate 

hirrde guardian 

hire her 

hirne corner 

hope Aq^g 

irre ire 

karrte cart 

kemmpe champion 

kene keen 

kide /;26? 

kinde hind s. 

kineriche kingdom 

kirrke church 

kirrkedure church- 

lade guiding s. 

Isechef ^^ccA 

l0efe belief 

lare ^re 

late, latef appear- 

lattre ?«<^«r 

lawe mound 

lajhe t ^«w 

lefe leave 

leode people 

leome, leme gleam 

le^he w^a^^s 

lifisshe living 

like /orm 

liref /oss 

li])e ^«YA^ gentle 

lojhe t ^re 

lufe /ot^g 

macche, make, 
mate, wife 

meenef company 

male f tribute 

mare more 

maj^stre master 

maj3}?e ifn'^^ A:w 

mede f meed 

mele mea^ 

merrke f mark 

messe mass 

mete meat 

rci^-^Q female cousin 

milde wzYc? 

milef mile 

millce mercy 

minde f mind 

minnstre f minster 

missdedef misdeed 

mone moon 

name name 

naj^e f ^race 

neddre adder 

nedlef needle 

-nesse -ness 

nesshe soft 

orrmete measureless 

orrtrowwe distrust- 

orrtrowwj^e distrust 

oxe ox 

pappe f breast 

pine pain 

profe'te prophet 

resste repose 

riche kingdoms rich 

rime f metre 

rodef rood 

rote root 

rume wide 

rune counsel 

ssete t sm^ 

sahhte concurring 

sake dispute 

sallfe sfl/t;^ 

sallmef psalm 

same f same 

sawle 5oe</ 

scone beauteous 

seollj^e sellj^e Ac^- 

serrjhe sorrow 

sexe s?:r 

sexte sixth 

sextene sixteen 

shsejief sheath 

shaffte creature 

shame shame 

shande disgrace 

shene sheen a. 

shriffte shrift 

sihhjje sight 

sijef victory 
smere ointment 
smej^e smooth 
soffte soffte 
spseche speech 
stede stead place 
staff ne voice 
steorrne star 
stirne stern a. 
stoke t stock 
strsete f street 
strande f strand 
strenncj^e strength 
sune son 
sunne sun 
sware f answer s. 

grievous a. 
swepe whip 
swij^e great 
tale tale number 
temmple temple 
tende tenth 
tene ten^ injury s. 
time time 
tunge tongue 
turrtle turtle 
twinne tivin 
]7eode people 
j^e^jre their 
j^ra^hef t?irow,iime 
J^ridde third 
])rinne three 
)7rittene thirteen 
J^rittennde thir 

J^rowwinngef throe 
jjurrfe needful 
))usennde thousand 
unnclene unclean 
unncweme unac- 
imnffele deceitful 
unnfsewe not a few 
unnhselef unsound- 
unnorne plain 
unnride vast 
unnsmej^e uneven 
mmwine enemy 

unnwrseste weak 

uppbrixle object of 

ure our 

W8ede clothing 

wsedle poor 

wsete t drink s. 

waldef power 

wambe belly 

wasstme fruit 

wajhe wall 

wecche watching s, 

wehhte f weight 

were were man 

werre worse 

wersse worse 

wesste waste desert 
s. and a, 

wej^e way 

whgete wheat 

wicke mean weak 

widdwe widow 

wilde wild 

wille will 

WIS, wise ivise a. 

wise wise s. 

wite prophet 

wite t punishment 

witejhunnge pro- 

wi]?]?errstrennc]7e f 
opposing power 

wlite t face 

wraeche vengeance 

wra]7]?e t wrath s, 

wrecche wretched 

wrihhte (1) maker; 
(2) blame 

wnde wood s. 

wnke week 

wullet wool 

wnnde f wound 

wurrj^ef worship 

wurrjjshipe worship 

^ate gate door 

jerrde f yard rod 

jife gift 

jure your 

It will be found on examination tliat thongli many of the above 
-e are justified by the existence of some final vowel or syllable in 
Anglosaxon or Icelandic, not a few have been clearly subsequently 
developed. See supra, p. 345, note 2, and the Table, pp. 379-397, 



Chap. V. 

2. Lajamon's Brut, beginning of xiiith Century. 

Although Lajamons Brut^ is written in verse, yet the rhythm 
and orthography are so irregular that it is scarcely easier to con- 
jecture the pronunciation than if it were mere prose. In fact with 
Orrmin we take leave of all certainty arising from metre or strict 
orthography. But the extraordinary diversity of spelling is of 
itself some assistance. 

Weighing the results already obtained we cannot be very far 
wrong in supposing a, e^ i, o, u to be (aa a, ee e, ii i, oo o, uu u), 
with the doubtful (i) or (y) for u occasionally as in lut, lutel, Inhere 
(h't, Kt'el, h'dh'cre) few, little, wicked.'^ Again <^ may be called 
(ee, e), and as eo interchanges with e it may be (ee) or (ee|_o). 
Ea is rare and interchanges with a, so that it may be (ea) or even 
(ea) with a more distinct (a). Among the consonants j, h, follow 
the same rule as in Orrmin, ch is of course (tsh), but (sh) does 
not seem to have been developed, as sc is constantly used. 

On account of the extreme western locality of the author's resi- 
dence (3 J miles south-east of Bewdley, in Worcestershire) there 
may have been many dialectic peculiarities which would tend to 
give the letters slightly different values from those thus assigned, 
but it seems probable that such a pronunciation as the following 
would have been intelligible.^ 

Laj^amorCs Brut. 

Madden' s edition, vol. i. p. 124, v. 2922. 

Sixti winter hefde Leir -^ 
fis lond al to welden. 
fe king hefde free dohtren 'i 
bi his drihliche quen. 
nefde he nenne sune ^ 
J7er fore he war^ sari, 
his manscipe to halden i 
buten Ja ]7reo dohtren. 

^ Lajamons Brut, or Chronicle of 
Britain ; a poetical semi-saxon para- 
phrase of the Brut of "Wace, now first 
pubhshed from the Cottonian manu- 
scripts in the British Museum, accom- 
panied by a hteral translation, notes, 
and a grammatical glossary. By Sir 
Frederic Madden, K.H., keeper of the 
MSS. in the British Museum. Pub- 
lished by the Society of Antiquaries of 
London, 1847, 3 vols, royal 8vo. The 
Cottonian MSS. are Calig. A. ix, the 
older version, which is attributed to the 
beginning of the xiiith century at 
latest, and Otho. C. xiii, which is of a 
much later date. 

Conjectured Pronunciation. 

Siks'ti win'ter neevde Lair 
dhis lond al to-weld*en. 
Dhe kiq neevde three dokht'ren 
bii nis driAh'litshe kween. 
Neevd He nen*e suuu'e, 
dheerfoor* He wardh sar'i, 
His man'skiipe to hald'en, 
buut'en dha three dokht'ren. 

2 The forms litul, li^ere also occur. 
It is quite possible that in such words 
both modes of speech (lut-el, lit-el) oc- 
curred in these Western dialects, see 
p. 298, p. 300 note 2, and p. 424. 

3 The many interesting points which 
would arise from a careful study of the 
dialectic peculiarities indicated by the 
orthography are of course passed over 
here, as the object is only to ascertain 
the phonetic meaning of the letters, 
which is an entirely preliminary inves- 
tigation without which the other could 
not properly succeed, but which is quite 
independent of any other research. 

§ 2, No. 2. 



fa aeldeste dohter haihte Gor- 

fa o^er Eegau. 
fa fridde Cordoille. 
Heo wes fa jmigeste suster ^ 
a wliten aire uairest ; 
heo wes hire fader al swa leof ^' 

swa his ajene lif. 

fa aeldede f e king t* 

& wakede an a^elan. 

& he hine bi-f ohte :! 

wet he don mahte. 

of his kineriche f* 

sefter his deie. 

He seide to himsuluen :! 

fat fat vuel wes : 

Ic wile mine riche to-don :! 

& allew minen dohtren. 

& 3 en en hem mine kine-feode r' 

& twemen mine beamen. 
Ac serst ic wille fondien t' 
whnlchere beo mi beste freond. 
and heo seal habbe fat beste del :! 

of mine drihlichen Ion. 
f us f e king f ohte :! 
and f er aefter he worhte. 

Dha Eld-este dokh'ter nai^hte 

dha oo'dher Reeg'au 
dha thrid'e Korduil'e 
Heo wes dha jnq*este sns'ter, 
a le^;ii-ten al*re vair"est. 
Heo wes niir'e faa'der al swa 

swaa His aagh'ene liif. 
Dhaa Eld'ede dhe kiq 
and waa'kede an aa'dhelan 
and nee niin'e bithokht'e 
whet He doon makht'e 
of his kin'eriitshe 
Eft'er His dai"e. 
He said'e to him sel'ven, 
dhat' dhat iivel wes : 
Ik wil'e miin*e riitsh'e to-doon 
and alien miin*en dokht'ren, 
and jeeven Hem miin'e kin*e- 

and tweem*en miin'e beam 'en, 
ak EErst ik wil'e fond'jen 
whilk'ere beo mi best'c freond, 
and Heo skal nab'e dhat best'e 

Of miin'e dri^h'litshen loon, [deel 
Dhus dhe kiq thokht'e 
and dheeraft'er He workht'e. 

Sir F. Madden' s translation of the above, omitting the parts relating to the 

more modern text. 

Sixty winters had Leir 

this land ' all ' to govern. 

The king had three daughters 

by his noble queen ; 

he had no son, — 

therefore he was sorry, — 

his honor to hold, 

except the three daughters. 

The eldest daughter hight Gornoillej 

the second Regan, 

the third Cordoille. 

She was the youngest * sister,' 

of beauty fairest of all ; 

she was to her father as dear 

as his own life ! 

Then the king grew old, 

and weakened in strength, 

and he bethought him 

what he might do 

with his kingdom, 

after his day. 

He said to himself 

that that was evil : 

" I will divide my realm 

to ' all ' my daughters, 

' and give them my kingdom, 

and share among my children ; ' 

but first I will prove 

which is my best friend, 

and she shall have the best part 

of my lordly land." 

Thus the king thought, 

and thereafter he wrought. 



§ 3. Prose Writings of the xiii th Century and Earlier. 

Here we have only the spelling to trust to, and to see 
whether the determination of the values of the letters by 
means of the poets is borne out by the systematic ortho- 
graphy of the prose writers. Very brief notices are all that 
need to be given. 

1. Only English Proclamation of Henry III, 18 Oct. 1258. 

This proclamation, issued by the barons in the king's name, has 
been fully considered in a separate work,^ in which the pronuncia- 
tion was assigned in accordance with the results at wluch I had 
then arrived,^ but subsequent research has induced me slightly to 
alter my opinion on certain points. Considering that the document 
is formal, it seems probably that ea, eo had their full (ea, eo) sounds. 
It is even possible that eow may have been (eou) rather than (eu), 
but the constant practice of writing ew in trewe leads me to believe 
that the initial eo of this combination has to be read (e) simply. 
The occurrence of simple ew, however, casts some doubt upon this 
conclusion as respects the actual pronunciation of the scribe. There 
is probably little doubt that the more general pronunciation of ea^ 
eo, at that time was (ee), and of eow (eu). The combination oa is 
rare. We have seen it rhyme with (aa) in Genesis and Exodus 
(p. 467), and the vmter may have said (aa, aa, aah), the last as an 
intermediate sound. As a compromise I use {aa, a). The inter- 
change of ce, e m rcedesmen redesmen, seems to imply that ce had 
become simple (ee, e). In accordance with former usage (ai) is 
employed for ei_ but we must not fail to obseiwe the correspondence 
of the French Fi% Geffrey, p. 504, with the English Geffrees sune 
p. 505, shewing that the pronunciation (Dzhef'ree*) was then 
current (supra p. 462). The name AlditheV in the Enghsh, p. 504, 
and AuditJieV in the French, p. 505, seems to be a contraction for 
the nscme Aldtdeleye in Staffordshire (Domesday Book, printed edition, 
fo. 250b, col. 2, photozincographed edition, Staffordshire, p. x. col. 2,) 
—ald-ide-leye, or ags. eald y^a lega, that is, old-waier-land, com- 
pare Caedmon's ea-stream-y'^a. Ide, still called (lid) supra p. 291, 
is in Devonshire (Domesday Book, fo. 101 J, col. 2,) as also Ideford; 
Idehill is in Kent, Iden in Sussex. Hence the probable alteration 
of the name was (ald-ii'dha-lee"^ha, ald-ii*dhe-lai, auld-i-lai, 
aud-e-lai, AAd'lee, AAd'K), compare Audelay, p. 449, n. 2, and the 
modern Audley. The other vowels and the consonants present no 
difficulty. The length of the vowels, where it differs in my scheme 

^ The only English Proclamation of * The error of supposing long i to 

Henry III, 18 October 1258, and its have been occasionally (ai), see supra 

treatment by former editors and trans- p. 279, was not detected till after the 

lators, considered and illustrated; to book had been printed off, and is re- 

which are added editions of the Cuckoo ferred to in the errata. The use of 

Song and The Prisoner's Prayer, Lyrics Henr' .... send igretinge for 

of the XIII th century, London, 1 868, s e n d e ]?, is well illustrated by Prof. 

8vo. pp. 135, by the author of this F. J. Child, supra p. 354, art. 61. 

§ 3, No. 1. HENRY III. XIII TH CENTURY. 499 

from that assigned to Anglosaxon, will generally be found justified 
by the spelling of Orrmin, or by more recent usage. The quantity 
of the Anglosaxon short vowels seems to have frequently suffered 
in passing through the l^Torman period of repression, when the 
language ceased to be cultivated by men of letters. 
■- The complete proclamation, with the French original, is here 
reproduced from the stereotype plates of the work cited in note 1, 
in order that the first correct presentation of this venerable and 
interesting document may be preserved for the use of the Early 
English Text Society. To insure accuracy, tj.e proofs had been 
compared three times with the originals in the Pi-blic Record Office. 
A few very slight inaccuracies in the stereotype plates have been 
removed in this edition, after a fourth comparison. The bracketed 
numbers refer to the numbers of the lines in the original MSS. 

The following is an abstract of the history of this important pro- 
clamation, the only public EngKsh document known to have been 
issued under our Norman kings. On account of the quarrels be- 
tween Henry III. and his barons, the latter were summoned to 
Westminster 7 April, 1258, when Henry submitted himseK to a 
Council of Twenty-four, twelve chosen by himself, and twelve by 
the Barons, or, as they called themselves, the Commons. This 
Council appointed a Committee of Four to choose a Cabinet of 
Eifteen. To this Council and Cabinet were due the provisions of 
Oxford, 11 June 1258, which ordered a Parliament consisting of the 
Fifteen, and Twelve Magnates to meet three times a year, and for 
the first time on 6 October 1258. At this Parliament the follow- 
ing Proclamation was agreed to, and issued in Latin, French, and 
English. The Latin version has not yet been found. There are 
two copies of the French, and one of the English in existence. 
The French version which follows contains the names of thirteen 
out of the Cabinet of Fifteen, and three from among the first ap- 
pointed Twelve Parliamentary Magnates. The object of the Pro- 
clamation, was to make each man in the country take the oath 
already taken by the King and the Commons at Oxford, pledging 
him to obey the Council of Twenty-foirr, to assist it to the utmost 
of his power, and to oppose its enemies. 

The English proclamation seems to have been published from the 
original by Somner 1659, Hearne 1720, Henshall 1798, the Eecord 
Commission (in its edition of Rymer's Foedera 1816,) the Master of 
the Rolls (in Sir H. James' photozincographed National Manuscripts 
1865), and, in part, by Astle 1803 (in facsimile), but in all cases 
incorrectly, and the errors made by these editors have increased 
in the hands of Tyrrel I700, Lyttelton 1767, Henry 1781-93, 
Latham 1841, and Koch 1863, who followed Somner ; and Craik 
1851, who followed Rymer. Pauli 1853, and Regel 1856 (who is 
followed by Marsh 1862,) conjecturally, and on the whole satis- 
factorily, amended Rymer by means of the French version, which 
has been published by Rymer and Pauli only, but the latter merely 
transcribed the former, leaving a grievous blunder uncorrected. 
Some of the errors of these various editions are given on page 504. 


Patent Roily 42 Henrij III. m, 1, w. 1. 

[1] Henri par la grace deu Eey de Englet're Sire de 
Irlande. Due de Normandie de Aqui'en et Cunte de Angou. a 
tuz fes feaus Clers et Lays saluz. Sachez ke nuf uolons et 
otrions ke ce ke noftre conseil [2] v la greignure partie de 
eus ki est esluz par nuf et par le co'mun de noftre Beaume a 
fet V fera al honur de deu et noftre fei et pur le p'fit de noftre 
Reaume ficum il ordenera :! feit ferm et eftable [3] en tuttef 
cliosef a tuz iurz. Et comandons et enioinons a tuz noz 
feaus et leans en la fei kil nus dement kil fermement teignent 
et iurgent a tenir et a maintenir les eftabliffemenz [4] ke funt 
fet V funt a fere par lauant dit Cunseil v la 

Modern English Translation of Old JEnglish Version, 

[1] Henry, by tbe grace of Grod, king of England, 
Lord of Ireland, Duke of Normandy, of Aquitaine, and 
Earl of Anjou, sends greetings to all his lieges, clerical and 
lay, in Huntingdonshire. [2] That know ye well all, that 
we will and grant that that which our councillors, all or 
the greater part of them, that have been chosen by us, 
and by the people of the country of our kingdom, have 
done, and shall [3] do, to the glory of God, and in fur- 
therance of our allegiance, for the benefit of the country, 
by the provision of the aforesaid councillors, be stedfast and 
lasting in all things ever without end. And we call upon 
[4] all our lieges in the allegiance that they owe us, that 
they stedfastly hold and swear to hold and to defend the 

'Zyr^ or mi^ that have been passed, or shall be passed by the 
jsf' ^ aforesaid ^^puncillors, [5] or by the 



§ 3, No. 1. HENRY III. — XIII TH CENTURY. 601 


Patent Roll, 43 Senry III. m. 15., n. 40 

[1] ^ Henr* J^urj godef fultume king on Engleneloande. 
Lhoauerd on Yrloand'. Duk on Norm' on Aquitain' and 
eorl on Aniow Send igretmge to alle hife holde ilaerde 
and ileawede on HuntendonTchir' [2] ])88t witen je 
wel alle J^aet ive willen and vnne?^ ])8et. J^aet vre rsDdef- 
men alle oj'er ]?e moare dael of heom J?8et beoj? icho- 
fen ])ur5 uf and furj J'aet loandef folk on vre 
kuneriche. habbej) idon and fchuUew [3] don In ]?e 
worJ^nefTe of gode and on vre treowj^e. for fe freme 
of ]>Q loande. J^urj ]>& befijte of J'an to forenifeide 
redefmen :! beo fledefseft and ileftmde in alle ])inge 
abuten sende. And we boaten [4] alle vre treowe in 
\e treow]?e J'aet beo vf ojen. ]?8et beo ftedefseftlicbe 
healden and fwerien to bealden and to werien ]>o 
ifetnefTef }?a6t beon imakede and beon to makien J^urj 
{)an to foren ifeide raedefmen [5] oj'er Jurj Je 

Conjectured Pronunciation of Old English Version, 

[1] Hen'rii tburkb God'es ful'tume kiq on Eqienel«n*de, 
Ibaverd on Iiri<2nde, Dyyk on Normandii, on Akitain'e and 
eorl on Andzbuu*, send igreet'iqe to al*e nis'e Hold'e ileer'de 
and ilee'wede on Hun*tendoonesbii*re. [2] Dbet wii'ten Je 
wel al*e, dbet we wil'en and un*en dbet, dbet uu're ree'des- 
men al'e odb*er dbe maa'TQ deel of Heom, dbet beotb itsboo*- 
sen tburkb us, and tburkb dbet l«nd*es folk on uu're 
kin'eriitsbe, Hab'etb idoon* and sbul'en [3] doon, in dbe 
wortb'nese of God'e and on uu're treutb'e, for dbe free'me 
of dbe land'e, tburkb dbe besi/i;b*te of tban to foo'renisaide 
ree'desmen, beo stee'defest and jles'tinde in al'e tbiq*e 
abuut'en en'de. And we naa'ton. [4] al*e uu're treu'e in 
dbe treutb-e dbet Heo us oogb'en, dbet Heo stee'defestliitsbe 
Heald'en and swee'rien to Heald'en and to weer'ien dbo 
iset'neses dbet beon imaa'kede and beon to maak'ien tburkb 
dban to foo'ren isaid'e ree'desmen, [5] odb'er tburkb dbe 


Old French Version, — (Continued.) 

greignure partie de eus. en la maniere kil est dit defuz. 
et kil fentreeident a ce fere par meifmes tel s'ment 
cunt' tutte genz [5] dreit fefant et p'nant, et ke nul 
ne preigne de t're ne de moeble par quel cefte purueance 
puifTe eftre defturbee v empiree en nule manere. et fe 
nul V nus viegnent encunt' cefte cbose [6] nuf uolons 
et comandons ke tuz nof feaus et leans le teignent a enemi 
mortel. et pur ce ke nus volons ke ceste chose feit ferme et 
eftable :! nof enueons nof lettres ou'tes feelees de n're [7] seel 
en cbefcun Cunte a demorer la entrefor. Tesmoln Meimeifmes 
a Londres le Difutime lur de Octobre Ian de noftre regne 
Q'raunte fecund. Et cefte cbose fu fete deuant Boneface 
Arce[8]eueske de Cantrebur'. Gaut' de Cantelou. Eueske de 
Wyreceftr'. Simo?^ de Montfort. Cunte de Leyceftr*. Richard 
de Clare Cunte de Glouceftr' et de Hertford. E-og^ 

Modern English Translation of Old English Version, — (Con.) 

greater part of them, as it has been before said. And that 
each help the other so to do by that same oath, against all 
men, doing and receiving justice. And let no man take 
any land or [6] chattel, whereby this provision may be 
let or impaired in any wise. And if any person or persons 
oppose this provision, we will and enjoin that all our lieges 
hold them as mortal enemies. And because [7] we will 
that this should be stedfast and lasting, we send you this 
letter patent signed with our seal, to hold among you in 
the treasury. Witnesses ourselves at London, the eigh- 
teenth day of the month [8] of October, in the two and 
fortieth year of our reign. And this was done in the 
presence of our sworn councillors, Boneface, archbishop of 
Canterbury ; Walter of Cantelow, bishop of Worcester ; 
[9] Simon de Montfort, earl of Leicester ; Eichard of Clare, 
earl of Gloucester and Hertford; Boger 

§ 3, No. 1. HENRY III. XIII TH CENTURY. 503 

Old English Version, — (Continued.) 

moare dsel of heom alfwo alfe hit if biforen ifeid. And 
>£et 2d\ic o>er helpe >3et for to done bi >an ilcbe 
o>e ajenef alle men. Ei^t for to done and to foangen. 
And noan ne nime of loande ne of [6] ejte. wherj^urj 
]?if befijte muje beon ilet o]?er iwerfed on onie 
wife. And jif oni o)er onie cumen Iter onjenef:^ 
we willen and hoaten j^aet alle vre treowe beom healden 
deadhche ifoan. And for >8et [7] we willen >8et >if beo 
ftedefgeft and leftmde r^ we fenden jew >if writ open 
ifeined wi]? vre feel, to balden a mangef jew inebord. 
WitnefTe yf feluen set Lunden'. >ane E5teten>e day. 
on >e Mon>e [8] of Octobr' In >e Twoandfowerti5>e 
jeare of vre cruninge. And \\( wef idon setforen 
vre ifworene redefmen. Bonefac' Arcbebifchop on Kant'- 
bur\ Walt* of Cantelow. Bifcbop on Wirecbeftr'. [9] 
Sim' of Muntfort. Eorl on Leircheflr'. Eic' of 

Clar* eorl on Glowcbestr' and on Hurtford.' Bog' 

Conjectured Pronunciation of Old English Version. — (Con.) 
"maa'VQ deel of Heom al'swo als'e nit iz bifoo'ren isaid*. And 
dbet eetsb oodb'er Help'e dbet for to doon'e bii dhaan il'tsbe 
ootb'e ajee'nes al'e men, likhi for to doon'e and to faq*en. 
And n^^n ne nii'me of It^nd'e ne of [6] e/tbt*e, wbeerthurkb* 
dbis besi/cbt'e muugb'e beon ilet* odb*er iwers'ed on on*ie 
wiise. And jif on'i odh'er on*ie kuum'en Heer onjee'nes, 
we wil'en and H^^'ten dbet al*e uu're treu'e Heom Heald'en 
deadiitsbe iiaan' And for dbet [7] we wil'en dbet dbis beo 
stee'defest and lest'inde, we send'en Jeu dbis rz^it oop*en 
isain'ed with uu're seel, to nald'en amaq'es Jeu in*e Hoord. 
Wit'nese us selven et Lun'deene, dbaan'e e^^bt'etentbe dai, 
on dbe moontb'e [8] of Oktoo'ber in dbe twoo and foour*ti/t:btbe 
jear'e of uu're kruun'iqe. And dbis wes idoon* etfoo'ren 
terber'i ; Walt'er of Kan'teloou, bisb'op on Wii'retsbester ; [9] 
Sii'moon of Munt'fort, eorl on Lair'tsbester ; E-ii'tsbard of 
Klaa're, eorl on Gloou'tsbester and on Hert'ford ; Eodzb'er 


Old French Version. — (Continued.) 

le Bigod Cunte de [9] Norf et Marefclial de Englet're 
Humfrey de Bohun Cunte de Hereford. Piere de Saueye. 
Gruilame de forz. Cunte de Aubemarle. lohan de Plesseiz 
Cunte de Warrewyk*. -Elog' de Quency [10] Cunte de 
Wynceftr'. lohan le Fiz Geffrey. Piere de Muntfort. 
Eicliard de Grey Bog* de Mortemer lames de Audithel. 
et Hug' le Despens*. 

Modern English Translation of Old English Version. — (Con.) 

Bigod, earl of Norfolk and Marshal of England : Peter of 
Savoy; William de Fort, earl of Albemarle; [10] John de 
Plessis, earl of Warwick ; John Fitz Geoffrey ; Peter de 
Montfort; Eichard de Grey; Roger de Mortimer; James 
Audley, and in the presence of other kinsmen. 

[11] And in exactly the same words it has been sent 
into every other shire throughout the kingdom of England 
and also in till Ireland. 

P rincipal errors of former editions. Only such blunders are here 
given as make nonsense of the original. The numbers refer to the 
lines of the MS., the spaced letters to the original, and the italics 
to the errors. 

Send igretinge 1, Tyrrel, Henry, Latham : send I greting. 

holde ilaerdel, HenshaU : hoi theilaerde. 

freme 3, Somner: freine ; HenshaU /mm(?. 

ilche o\q 5, Tyrrel, Henry, Lyttelton : ilche other. 

E. i J t 5, Somner : {in alle \inge ]><;et) ogt ; Tyrrel : {in all thinge 

thcet) ogt ; Henry, Lyttelton ; in alle thet heo ogt ; Craik : 

[in alle thaet heo'] oght. 
noan ne nime of 5, Somner, Tyrrel, Henry, Henshall, 

Eymer, Craik : noan ne mine of; Latham : noan ne of mine. 
ejte. wher]7ur5 6, Somner : egtewhcer \urg ; Tyrrel, Henry : 

egeteivher thurg ; Latham : egetewhere, thurg ; Henshall : 

egte-whcer^ thurg ; Eymer : egteohcero \urg ; Craik : eghteo- 

hcero, thurg. 
deadliche ifoan6, Somner, Tyrrel, Henry, Latham, Itymer, 

Craik : deadlicheistan; ; Henshall : deadliche. If than. 
In consequence of these errors the translations given by Somner, 

§ 3, No. 1. HENRY III. — XIII TH CENTURY. 505 

Old English Version. — (Continued.) 

Bigod eorl on Northfolk' and Marefcal on Engleneloand.' 
Perref of Sauueye. Will' of Fort eorl on Aubem'. 
[10] loh' of Plefleiz. eorl on Warewik loh' 
Gfeffreef fune. Perref of Muntfort. Ric' of Grey. Eog' 
of Mortemer. lamef of Aldithel' and setforen oj^re 

[11] ^ And al on ]>o ilclie worden if ifend in to seurilice 
oj^re slicire oner al J^sere kuneriche on Engleneloande. 
and ek in tel Irelonde. 

Conjectured Pronunciation of Old English Version. — (Con.) 
Bii'god, eorl on Nortli'folke and Maa'reskal on Eq-leneknd'e. 
Per-es of Savai'e; Wil'Helm of Fort, eorl on Au'bemarle; 
[10] Dzhoon of Ples'aiz, eorl on Waa'rewiike; Dzhoon 
Dzhefrees suun'e ; Per-es of Munt'fort ; Eii'tshard of Grai ; 
Rodzh*er of Mor'temer; Dzhaam-ez of Al'dithel, and etfoo'ren 
oodli're moogh'e 

[11] And al on dho il'tslie word'en is isend* in to evritslie 
oodh-re sliii're oo'ver al dhee're kin'eriitshe on Eq-leneknde, 
and eek in til lir'knde. 

Henry, Latham, and Craik of the passage : And faet aehc ofer 
helpe .... deadliche ifoan, 5, 6, are ludicrously wrong. 

Somner's Latin version is : *' Et quod unusquisque, vigore ejus- 
dem juramenti, contra onines homines, in omnibus tum faciendis, 
tum recipiendis, ut id ita fiat et observetur, alter alteri sint auxilio. 
Et (quod) nuUus sive de terra (vel, gente) mea, sive quacunque alia, 
per consilium hujusmodi (hujus scil. consilii obeundi causa) impe- 
diatur, sive damnum patiatur, uUo modo. Et si quis, sive vir sive 
faemina, huic (edicto) contravenerit, volumus et mandamus ut omnes 
fideles nostri eos habeant infensissimos." 

Craik' s English version is : " And that each other help that for 
to do, by them (to) each other against all men (in all that they) 
ought for to do and to promote. And none, nor of my land nor 
elsewhere, through this business may be let (hindered) or damaged 
in anywise. And if any man or any woman come them against, 
we will and enjoin that all our lieges them hold deadly foes." 

The most remarkable error in the copy of the Erench version 
printed in Rymer is : nos Gkceons, for nos enueons 6, which 
has the false appearance of an appropriation of a Saxon word by 
the Normans, with a Erench inflexion, — a philological curiosity ! 




Chap. V. 

2. Ancren Riwle, XIII th Century. 

The Anceen Riwle and the Halt Meidenhad may be considered 
together. ^ 

In the Anceen Eiwle it will be seen that the simple vowels 
a, e, i, 0, 21 must be taken as usual to mean (aa a, ee e, ii i, oo o, 
uu u), with a much larger allowance of m = (y) or (i, e) than is 
found, except in the west of England. Thus we have ffult, cluppen, 
fustes, fur, lupes, lut, mile, for guilt, clip (embrace), fists, fire, lips, 
little, n'ill. Besides this there is a very extensive assortment of 
diphthongs and even triphthongs, which should be apparently pro- 
nounced thus : ai, au, ea, ei, eo, eu, oa, oi, ou, ui = (ai, au, eea ea, 
ai, eeo eo, eu, ooa, uui, oou ou, ui). The oa, oi, ui as in hloawen 
lloamen luine are too rare to form a good judgment on. 

The combination iw which only occurs in the foreign word riwl 
is most probably intended to give the sound (yy), for it is scarcely 
possible to imagine that (yy) could not have been pronounced, and 
that therefore iw = (iu).^ On account of the action of the (r) the 
sound (riul) is difficult to enunciate purely, and (ruul, ryyl, rnl) 
are all easier, and they are consequently still in use provincially. 

The following brief example from p. 70 of the Ancren Eiwle, ^ 
will shew the efi'ect of these assumptions, and will render an ex- 
ample from Hali Meidenhad needless ; 

Original Text. 
Muche fol he were, fe 
muhte, to his owene bihoue, 
hwe^er se he wolde, grinden 
greet ojer hwete, gif he 
grunde J^et greet and lefde 
fene hwete. Hwete is holi 
speche, ase Seint Anselme 
sei^. Heo grint greet ^e 
chefle^. pe two cheoken 
beo'^ Je two grinstones. 
pe tunge is J^e cleppe. Loke^, 
leoue sustren, ]iet ouwer 
cheoken ne grinden neuer 

^ The Ancren Riwle ; a treatise on 
the Rules and Duties of Monastic Life, 
edited and translated from a Semi- 
Saxon MS. of the thirteenth century 
by James Morton, B.D., vicar of Hol- 
beach, prebendary of Lincoln ; printed 
for the Camden Society, 1853, London, 
4to. Kali Meidenhad^ from MS. Cott. 
Titus D. xviii, fol. 112 c, an allitera- 
tive homily of the thirteenth century, 
edited by Oswald Cockayne, M.A., 
London, 1866, 8vo. pp, viii, 50 ; 
E. E. T. S. 

2 As the combination iw does not 
occur in other words, and as riule, 

Conjectured Pronunciation, 
Mutsh'e fool Hee weer'e, dhe 
mukht'e, to nis oou'ene binoo've, 
whedh'er see ne wold'e, grind-en 
greeot oo'dher wheet'e, jif He 
grund'e dhet greeot and leevde 
dheen'e wheet'e. Wheet'e is HooLi 
speetsh'e, as'e Saint Anselm*e 
saith. Heeo grint greeot dhe 
tsheef'leth. Dhe twoo tsheek'en 
beoth dhe twoo grin'stoon'es.* 
Dhe tuq*e is the klep'e. Look*- 
eth, leo've sustren, dhet oou'er 
tsheok'en ne grind 'en never 

reule are found in very old Nor- 
man, the point must be considered 
doubtful. In the xiv th century the 
sound was almost certainly (ryy-le). 
Mr. Payne is inclined to think that the 
old Norman sound was (riu-le), 

3 The proof was read by Mr. Brock 
by the original MS., Cott. Nero A. xiv. 

* The "colloquial" pronunciation 
(grm-stan), mentioned by Smart, is 
thus shewn to be very ancient, and 
becomes a proof that grind was for- 
merly (grmd) not (graind), supra p. 
276, and p. 290, L 3. 

§ 3, No. 2 & 3. 



bute soule node i ne our 
earen ne hercnen neuer bute 
soule heale : and nout one 
our earen, aub ower eie 
furies tune^ ajein idel 
specbe f* ')^et to ou ne cume 
no tale, ne ti^inge of fe worlde. 


Much fool lie were, that might, to 
his own hehoof, whether so he would, 
grind chaff (grits) or wheat, if he 
ground the chaff and left the wheat. 
Wheat is holy speech, as Saint Anselm 
saith. She grinds chaff that chaffs 
(chatters). The two cheeks are the 
two grindstones. The tongue is the 

buut'e sooul'e vood'e ; ne oour 
ea'ren ne nerk'nen ner'er buut'e 
sooul'e neal'e ; and nout oon*e 
oour eea'ren, aukb oou'er ai'e 
tbirl'es tuuu'etb ajain* ii'del 
speetsh'e ; dbet to oou ne kuum*e 
ne taal'e netiidbiqe of dbe worlde. 


clapper. Look, dear sisters, that your 
cheeks do not grind never hut soul's 
food; nor your ears do not harken 
never hut to soul's health ; and not only 
your ears, but your eye's windows 
fence against idle speech ; (so) that to 
you (may) not come neither tale nor 
tiding of the world. 

3. Old English Homilies, xiith Century. 

Tbe venerable bomilies lately disinterred by Mr. Morris ^ cannot 
be read in any other way than the Ancren E-iwle. The values of 
all the letters and combinations seem to be completely known, and 
no further change can be expected. A very brief example will 
therefore suffice. In the following, the original text is exactly 
reproduced except in ^^'^mid for mi^^ ^'^^wolde for walde, ^^^ga?6 for 
gad, (*Wo^ for de^, ^%ulJce for huhe. The leinten for lenten at the 
beginning, may, as so many other evidently are, be a dialectic 
pronunciation, and is comparable with jieuh for fie8}i (supra p. 
473, n. 1), but Stratmann quotes the same form from Wright, 
Yocab. 90, Eob. Glouc. 187, 8. The experiment of writing (y) 
for w, when it may be (i, e), and (ei) for ei^ as being older forms, 
has here been made. 

Original Text, p. 25. 
Dominica Prima in Quadrigesima. 
[I]n leinten time uwilc mon 
ga^ to scrifte ; ]7er beo^ 
summe ]?e mare herm is fe 
ga^ al swa ic nu]7e eow tellen 
wulle. He sei^ mid'^'Jjamu^e 
fet nis naut in his heorte. ic 
wulle gan to scrifte for scome 
alswa do^ o^er men. ^if ic 
forlete fe preost me wolde^^^ 
eskien on ester dei hwa me 
serine er he me ^efe husul 
and ec for monne weordes 
^inge. he ne ga^ ^^' naut to 
scrifte al swa do^ o^er men. 

Conjectured Pronunciation. 
First Sunday in Lent. 

In lein-ten tii'me ywilk mon 
gaath to skrift'e. Dher beoth 
sum'e, dhe maa're nerm is dhe 
gaath, alswaa* ik nuu'dhe jou tel'en 
wyl*e. He saith mid dha muu'dhe, 
dhet nis naut in nis Heorte : "Ik 
wyl'e gaan to skrifte for skoo'me 
alswaa* dooth oo'dher men ; jif ik 
forlee'te, dhe preost me wol'de 
es'kien on eest'er dai whaa me 
skrii've, eer nee me jeewe hus'uI, 
and eek for mon'e weor'des 
thiq'e." Hee no gaath naut to 
skrifte alswaa* dooth oo'dher men. 

In the same work with the Moral Ode, supra p. 484, note 1. 



Chap. V. § 4. 

Ah al swa he do^^*| swa fe 
swica Jje biswike% hine 
seolfe on ende and bi^ al swa 
is an eppel iheowe^. he bi^ 
wi^uten feire and frakel 
wi^inne. Awah fet he efre 
wnlle J^ristelechen o^er bi- 
fenchcn mid his fule heorte 
fe heo wuUe underfon swa 
he J fing and swa hali swa is 
cr«stes licome in his sunfulle 
bulke.^^' and wene^ \et hit 
wulle him helpen :! J^eiso^- 
liche nawiht ah fenne fe 
preost hit de^ in his mufe. 
fenne cume^ drihtenes engel 
and binime^ ]7a halinesse mid 
him toward heouene riche. 
\et ]7er bilef ^ in his mu^e, 
ah ^if eni mon hit muste isean. 
he mahte iseon ane beminde 
glede \et hine al for-bema^ 
furut to cole. 

Akh alswaa* He dooth, swaa dhe 
swiik'a dhee biswii'keth niin'e 
seol'fe on end'e, and biith alswaa* 
is an ep'cl iheo'weth; nee biith 
withuu'ten fai-re, and frak'el 
within* e. Awakh-, dhet hb evre 
wyl"e thris'teletsh'en odh'er bi- 
then'tshen mid nis fuu'le neor'te, 
dhee neo wyl'e un'derfoon swaa 
Hei thiq and swaa naa'li, swaa is 
Ejist'es liic'oome in nis syn'fule 
bulk'e, and ween*eth dhet nit 
wyl'e mm Helpen ! I^ei, sooth*- 
liithshe naawikht ! akh dhen'e dhe 
preost nit deeth in iris muu'the, 
dhen*e kuu'meth drikhtenes eq*el, 
and binii'meth dha naa'lines'e mid 
Him toward* neo'vene rii*tshe. 
Dhet dher bilefth* in His muu'the, 
akh jif en*i mon nit mus'te isee*an, 
He makh'te isee'on aa*ne bem*ind*e 
gleed*e, dhet Hii*ne al forbem*eth 
thuniut* to koole. 

Mr, Morris's Translation, p. 24. 

In Lenten time each man goes to 
confession ; there are some to whom 
there is greater harm in going (than 
in abstaining), as I will now tell you. 
He saith with the mouth what is not 
in his heart. " I will go to shrift for 
shame, as other men do ; if I neglect 
the priest will ask me on Easter day 
who shrove me, before he administer to 
me the sacrament, and also for the sake 
of man's esteem." He does not go to 
shrift as other [good] men do, hut acts 
like the cheat who at last deceiveth 
himself, and is as a rosy apple— fair 

without and rotten within. Alas that 
he will ever dare or think with his foul 
heart to receive so high and so holy a 
thing as is Christ's flesh into his sinful 
body, and thinketh that it will help 
him. Nay truly not ! hut when the 
priest putteth it in his mouth, then 
cometh the Lord's angel and taketh 
the holiness with him toward heaven- 
kingdom. As for what remaineth there 
in ms mouth, if any man were able to 
perceive it, he might see a burning 
gleed that consumes him all to coals. 

§ 4. Teutonic and Scandinavian Sources of the English Language, 

The pronunciation of English has now been traced up to 
the earliest period in which it is known in a literary form as 
distinct from Anglosaxon. To complete the edifice, some 
account must be attempted of the pronunciation of Anglo- 
Saxon, the direct mother, and Old Norse, an important 
modifier of our tongue. These again point to Gothic as the 
oldest low German dialect that is known. It would be 
highly desirable to add an account of Old Norman, but no 

Chap. V. § 4. 



sufficient researclies have been made into that language to 
warrant any detailed statement of the pronunciation of that 
language. It must be therefore entirely passed over.^ 

^ See the observations on p. 438, and 
the remarks on Norman at, ei, ^. 453. 
Dr. Eapp, while owning that the de- 
cyphering of the phonetic meaning of 
Northern and Old French documents 
was one of the most difficult parts of 
his task, has yet ventured to assign 
such definite values to the symbols as 
to give detailed specimens, which he 
has not attempted for Anglosaxon and 
Old Norse. Although I am far from 
agreeing with his results, which appear 
to be founded upon insufficient exami- 
nation of the sources of information, 
the reader will probably be pleased to 
have a brief account of his opinions, 
Phys. d. Spr. ii, 82-117. The follow- 
ing seems to be ;his alphabet : A aa a, 
AI ee, AU au, B b, C k s sh, CH sh, 
I) d, E ee e d 9, 1E,I ei, EU oe ej, G 

Du Chevalier, qui ooit la Messe et 
Notre - Dame estoit pour lui au 

Extrait d'unMS. de SorbonneNo. 331 (2). 

Dous Jhesus, com cil bel guerroie, 

Et come noblement tournoie, 

Qui volentiers au monstier tourne, 

Oti Ten le saint servise atourne 4 

Et celebre le saint mistere 

Du doux Fils de la Vierge Mere. 

Pour ce vueil un conte retraire, 

Si com le truis en exemplaire. 8 

Un Chevalier courtois et sages, 

Hardis et de grant vasselages, 

Nus mieudres en Chevalerie, 

Moult amoit la Vierge Marie. 12 

Pour son barnage demener 

Et son franc cors d'armes pener, 

Aloit a son tournoiement, 

Garnis de son contenement. 16 

Au Dieu plesir ainsi avint. 

Que quant le jour du tournoi vint, 

II se hastoit de chevauchier : 

Eien vousist etre en champ premier 20 

D'une egiise qui pres estoit 

Oi' les sains que Ton sonoit 

Pour la sainte Messe chanter. 

Le Chevalier sans arrester 24 

S'en est ale droit a 1' egiise 

Pour escouter le Dieu servise, 

L'en chantoit tantost hautement 

Une Messe devotement 28 

De la sainte Vierge Marie, 

Puis a on autre comencie, 

g zh, H H, I ii, i, J zh, K k, L 
1, M m, N n, [AN aq, EN eq, IN 
iq, ON oq, UN yq, AIN EIN eq, 
OIN oiq, UIN uiq,] oo o o, (E ce, 
01 oi, OU u 9U, P p, Q k, E r, 
S s, T t, [-NT -n, -q], U y oe, UI 
■Qi, V V, X u s us ks, Y i, Z s ts. 
The following is a small portion 
of his example taken from Mienne 
Barbazan, Fabliaux et Contes des 
Poetes fran^ois des xi, xii, xiii, xiv 
et XV siecles, 1808, 8vo. 4 vols., vol. 1, 
p. 82, the original text, which Eapp 
omits, is here added by way of com- 
parison. As I have not been quite 
able to appreciate his system of accen- 
tuation, I omit it altogether. I have 
also forborne to correct any apparent 
errors, such as making meisme v. 35, of 
two, instead of three syllables. 

Dy shevali^er, ki o,oit la mese e 
notra dama estoit puur lui au 

Duus Zhesys, kom shil bel geroia 

E koma noblameq turnoia 

Ki voleqtiers au monstier turna, 

Uu 1- eq h seq servis- aturna 

E shdebra h seq mistera 

Dy duus Fils de la Vierzhe Mera, 

Puur she vcelj yq koqta r<?treera, 

Si kom h triris en eseqpleera.— 

Yq shevalier kurtois e sazhas, 

Hardiis e de graq vaselazhas, 

Nyys mieudres eq shevalariia, 

Mult amoit la Vierzha Mariia 

Puur soq barnazha demeneer 

E soq fraqk koors d- armas pmeer 

Aloit a soq turnoiameq 

Garnis de soq koqtenameq 

Au Dice plesiir eqsi aviq. 

Ke kaq h zhuur dy turnoi viq, 

II se Hastoit de shevaushieer ; 

Bieq vusit estr- eq shaq premieer. 

D- yn- eglisa ki prest estoit, 

0,ii los seqs ke 1- oq sonoit, 

Puur la seqta mesa shaqteer. 

Jj9 shevalieer saqs aresteer 

S- en est alee droit a 1- eglisa 

Puur eskuteer la Dice servisa ; 

L- eq shaqtoit taqtost Hautameq 

Yna mesa devotameq 

De la seqta Vierzha Mariia : 

Puis a on autra komeqshiia, 



Chap. V. 

1. Anglos AXON. 

The value of the letters in Anglosaxon proper could not have 
materially differed from that which the whole of the preceding in- 
vestigations has led us to assume for the letters used in the earlier 
part of the xm th and close of the xn th century. The most re- 
markahle difference was the vowel y, manifestly (yy, y), which 
however had hecome interchangeahle with i, and therefore equiva- 
lent to (ii, i) or (n, i) before the inflectional system of the Anglo- 
Saxon literature had disappeared. The vowel ce we may also 
assume to have had its deeper sound, now again familiar in England 
(aeee, se). It is very probable that a was sounded fully as broad as 
{aa, a), but e was probably not so broad as (ee e) because it would 
have been otherwise confused with (aeee, se). That short i was (e), 
from the Saxon times to the present day, there can be very little 
doubt, although, from having no direct authority for this conclusion, 
I have generally written it (i) before the xrv th century. But we 

Le Chevaher bien I'escouta, 

De bon cuer la Dame pria. 32 

Et quant la Messe fut finee, 

La tierce fu recomenciee 

Tantost en ee meisme lieu. 

Sire, pour la sainte char de Dieu, 36 

Ce li a dit son Escuier, 

L'heure passe de tournoier, 

Et vous que demourez ici ? 

Venez vous en, je vous en pri, 40 

Volez vous devenir hermite, 

Ou papelart, ou ypocrite ? 

Alons-en a nostre mestier. 

Amis, ce dist li Chevalier, 44 

Cn tournoie moult noblement, 

Qui le servise Dieu entenl, 

Quant les Messes seront trestoutes 

Dittes, s'en irons a nos routes : 48 

Se Dieu plest, ains n'en partirai, 

Et puis au Dieu plesir irai 

Tournoier viguereusement ; 

De ce ne tint parlement. 52 

Devers I'autel sa chiere tourney 

En saintes oroisons sejourne 

Tant que toutes chantees fnrent, 

Puis monterent, com fere durent, 56 

Et chevauchierent vers le leu 

Ou fere devoient leur geu. 

L? shevalie^ bieq 1- estuta, 
J)e boq koer la dama pria. 
E kaq la mesa fyt fin^^a 
La tiersha fy rekomeqshieea 
Tantost eq sh? meesma lioe. 
Siir, pur la seqta shar de Dioeoe, 
Shs li a dit son escuii^er, 
L- cera pasa de turnoieer, 
E vus \ee demurees ishii ? 
Venees vus eq, zh? vus eq prii, 
Voices vus dev6uiir nermita, 
TJ papalart u ipokrita ? 
Aloqs eq a nostra mesti^. 
Amiis, sh^ dist li shevalier, 
Shil turnoia mult noblameq 
Ki h servisa Dice eqteq ; 
Kaq las mesas seroq trestutas 
Ditas, s- en iroqs a nos rutas ; 
Se Dioeoe plest, eqs n- eq partiree, 
^puis a Dice plesir iree 
Turnoieer vigoeroesameq ; 
De sh? ne tiq parlameq. 
Devers 1- autel sa shiera tuma^ 
Eq seqtas oroisoqs sezhurna 
Taq ke tutas shaqteeas fyra, 
Puis moqtera, kom fera dyra, 
E shevaushiera vers h loeoe 
TJu fera d^voia loer zhceoe* 

3. monstier, monastere 
8. truis, trouve 
11. mieudres, meilleur 
13. barnage, courage, 
pener, tourmenter 
contenement, etat 



sains, cloches. 


26. le Dieu servise, le service de Dieu 

30. puis, on en a une autre commencee 

39. que, pourquoi demeurez-vous ici ? 

force, valeur, 42. papelart, faux devot 

43. alons-en, allons nous-en 

48. s'en irons, si nous, et nous nous en 


55. tant que, jusqu'd ce qm. 


find («') or even {e\ so rooted in tlie Korth of Europe at tlie present 
day, among not merely the English, hut the Scotch, Dutch, Danes, 
and Swedes, and ahove all, the Icelanders, who acknowledge it 
orthographically, that it presents the appearance of an original sound, 
rather than of a modern development. The o was almost certainly 
(oo o) ; the distinction {oo o) is quite of modem growth, nor have we 
been led to suppose that there was any equivalent distinction from 
the XVI th century upwards. The u was perhaps {uu u) rather than 
(uu u) or (uu w), the modem use. 

The digraphs ea, eo could scarcely have been (ja, jo) as Rask 
supposes, being misled apparently by modern Scandinavian usage. 
The confusions of ea with ce on the one hand, and a on the other, 
even in Angiosaxon, and its further confusion with e in more recent 
times, as the xni th century, exclude the sound of (j) with certainty.^ 
And similarly for (eo). But it is possible that they were occa^ 
sionally pronounced with the second element more conspicuous than 
the first, so that though we may generally write (ea, eo), as true 
diphthongs, in the ordinary manner, it may be occasionally neces- 
sary to indicate the preponderance of the second element by 
writing (ea, eo) or perhaps more truly (caa, eoo) which might fall 
into (aa, oo, uu). On examining the long list of Angiosaxon words 
commencing with ea eo, the following are all that I have noticed 
which could give rise to the notion of the pronunciation (ja jo), 
which Rask seems to have adopted through his own Scandinavian 
habits : ealo ale, vulgar (jel, j^l) : Eoforwic, in Domesday Eurvic, 
York, with the secondary form Eferwic ; eond yond, the proper form 
being geond, eow you; eowu ewe, dialectic (joo). Remembering 
how recently the sounds (w, j) have been prefixed to the English 
one, Scotch ane (won, jen), we can find no difficulty with these 
words. The Icelandic Jarl, which many persons rely upon for 
proving that ags. eorl must have been (jorl), was perhaps a deriva- 
tive of ar the hearth, and was anciently applied to an upper domes- 
tic, whereas the ags. word was probably connected with the old 
Saxon erl, constantly used for male, man, and in the plural erlosy 
and compound erlscepi for men, people,, collectively (Schmeller's 
Seliand, Gloss, p. 29). Hence the efi'ect of palatisation can alone 
be relied on in support of this (j) theory. 

I^ow the palatisation of a preceding c (k) into {h) would be produced 
by the simple action of the palatal (e) and would not require that 
that (e) should be squeezed into (i, j). Indeed, we have observed 
a tendency to palatisation in French and English before (a) sounds, 
which in French produced (kj, tsh, sh) (p. 53), but in English after 
flourishing for a little time as (ki, kj, kij) and still dragging out an 
obscure existence in a fast disappearing generation, or on the boards 
of second-rate theatres, (p. 206), is rapidly going out of use and 
favour.^ In modern French, too, both (kj) and (gj) are used with- 

1 The isolated identification of ea ^ j^ jg strongly marked in the dia- 

with (je) in certain words, by Sales- lects of the Peak of Derbyshire, 
bury, we have seen reason to suppose 
was a misprint, p. 80. 


out any tendency to becoming (sh, zh) as in queue, gueux (kjoe, 
gjoe). Icelandic is a conspicuous example of the same, as k, g 
are there always palatised into (kj, gj) before (aa/, ee, eei, i, 
i, j) without having the least tendency to become (tsh, dzh). The 
(k) itself is naturally an unstable letter ; cither the tongue has a 
tendency to rise, producing (kj), or the lips a tendency to round, 
producing (kw), and from these physiological actions can be traced 
a vast variety of changes in time and place. The same remarks 
apply also to (g) and to (kh, gh). A proper understanding of the 
relations, palatal (k, kj, tj, tsh, sh), and (kh, kjh, jh, j, i), labial 
(k, K, k«^, w, b, p) and (kh, xh, ke^h, wh, f ) will serve to solve 
numerous riddles in comparative philology. JS^ot only does, how- 
ever, a palatal vowel by direct action, or occasionally a guttural 
vowel by contrary action, tend to palatalize a consonant, but also 
the presence of the liquids (1, m, n, r) produces the same effect in 
the Germanic languages, as we have already had occasion to observe 
(p. 205). It is curious to note how certain words, however, resist 
palatalization, while their fellows readily succumb to the influence, 
as in drinh drench. The resistance to palatization is not purely 
Scotch. We find werchen in the Peisoker's Peayer, v. 41, and 
werch often in Chaucer, but we constantly find werk. In the 
Anceei^ Eiwle, while k had yielded to (tsh) by itself, sc had not 
become (sh), as in Italy and Germany, and as generally in England 
at that time, and the modem shot scot, ags. sceat, shews both the 
palatized and unpalatalized form of the same word still current. 
Again although cealc is now chalk (^ealk, tshAAk), and ceap is cheap 
(^eap, tsheep, tshiip), ceald, cealf are cold, caip (X*eald, k«ald, koold, 
koould, koold ; ^-ealf, \a(M, kaaulf, kauf, kaaf ), and if cicen has be- 
come chicken (tsh^k•en), altering the first and retaining the second 
(k), cicenehsiS become kitchen (kitsh' en)}) j a precisely contrary action. 
Again, the single word wicca seems to have given rise to both witch 
and wicked, (wicke in Orrminn) and similarly ags. wic gives wick as 
an independent word, also heard in Wickham and in teiminations as 
lailiwick, sherifwick, as well as Berwick, Alnwick, while in other 
cases it gives (w^tsh) as in Ipswich"^ or («dzh) as in Norwich. Hence 
the pure (k) is no more the sign of a north country pronunciation than 
the (tsh) of the south ; nor is it at all necessary to suppose that ea, eo 
were (ja, jo) to account for the change of a preceding (k) into (tsh). 
As to the consonants generally there is very little to observe, 
except that probably (kj, gj) were well in use in the early Anglo- 
saxon times, that g also probably became (^h) that is, (gjh) in many 
cases, in the same way as it now does in Iceland, and in Modem 
Saxony,^ so that the preparation for the (j) or simple (i) sound was 
early made. On the other hand, after (o, u) sounds and in other 

1 In Cumberland (koof). ^ Modern Saxon is high German, 

2 So called generally by persons old Saxon and Anglosaxon low German, 
living away from East Anglia. In There was no connection between the 
Norwich I heard it called (/ps*/dzh) two, and no connection is intended to 
which follows the analogy of Norwich be implied by this illustration. They 
and Greenwich. are two independent phenomena. 


places g may have had an early tendency to (g^^h) as we also find 
in Icelandic, and thus prepared the subsequent changes (p. 212 
and p. 311.) 

The letter h seems to have naturally played a triple part, the 
three functions being frequently confused, and by no means gene- 
rally understood at the present day. At the beginning of words h 
was either (h) or (h'), probably sometimes one and sometimes the 
other as in modern English, and in almost all languages where h is 
pronounced at all. At the end of words, the (h') was replaced by 
the (kh) which is an easier terminal sound, and more adapted to 
check a vowel sound. The initial combinations hi, hr, hn, hw, are 
ordinarily assumed to be (khl, khr, khn, khw) and at a remote 
period, before Anglosaxon was properly constituted, they may have 
been (kjhl, kjhr, kjhn, \.w\i). Eut it seems more probable that in 
the more cultivated period they were reduced to (Ih, rh, nh, wh), the 
last (wh) remaining to the present day, although sadly neglected in 
the South of England, and the first (Ih) existing in the xiii th century, 
though the second and third (rh, nh) rapidly disappeared. This 
view is strongly confirmed by the existent Icelandic pronunciation 
of hj, hi, hn, hr, hv as (jh, Ih, nh, rh, wh). The device of pre- 
fixing h to form the symbols for these sounds, is so natural, that 
many persons still insist that the proper way of writing when is hwen, 
and when I was printing phonetically I found tliis position of the 
letters practically sufficient. An accurate analysis, however, would 
shew that (n'wen) was materially diff'crent from (when), and that 
therefore in all accurate phonetic writing the sounds should be 

The letter p^ I presume was (w), certainly not (v), and probably 
not (bh). It is supposed by some to be merely a variety of the 
medieval form of v, but I consider it to be rather the old rune called 
wen = hope, in Cotton MS. Otho B. 10, as quoted in Hickes's 
Anglosaxon Grammar (Thesaurus i. 135). The sound of v con- 
sonant in ancient Latin, is a matter of dispute ; it was probably 
(w) or (bh), and more probably the latter than the former, because 
we can hardly imagine (w) generating (v) except through (bh), but 
the passage from (bh) to (v) is so easy and slight, that the two 
parts of Germany which are distinguished by the two different 
sounds at this day, profess to pronounce their w in the same way. 
(Bh) is a kind of bat sound, readily fallirig into (w) or (v), but the 
real (w) has a very moderate domain in Europe.^ The (bh) is 
thoroughly established in high Germany and in Spaia, where the 
old joke of 

"felices popuh qiiibus vivere est bibere " 

'^ Mr, Skeat notices only seven or at about 1300." — Havelock, Preface 

eight instances of the use of p in § 26. 

Havelock, adding: "This evidence is ^ ^^ accurate conception of the 

interesting as shewing that this letter three sounds (w, bh, v) is necessary for 

was then fast going out of use, and I the proper understanding of many 

think we may safely date the final dis- linguistic relations. For (w) the lips 

disappearance of this letter from MSS. are rounded nearly as for (u) and the 




Chap. V. 

points at once to the antit^iiity of the sound in that country in whicli 
it is still used for both I and v, and to the probable pronunciation of v 
in Latin as (bh) at that time. The example of Kavv6a<; being heard 
as cav^ rC eas = cave ne eas, would be solved by the identity (kabhne* 
aas) in both languages at that time. At the time when the Anglo- 
saxons, being Chiistianized, adopted the Christian Roman alphabet, 
the Roman v consonant was certainly (v), a sound which the Anglo- 
Saxons did not then distinguish from (f ), as we have reason to 
suppose that the letter /, like the letter s, served the purposes of 
both hiss and buzz. The consequence was that the Anglosaxons 
had no sign for their w consonant, which was distinct from v, and 
they therefore retained their runic p. For these reasons I think 
that p was (w) not (v), and that the German habit of transliterat- 
ing p by V is improper. 

The combinations cw, wl, wr, were probably the labial modifica- 
tions (kw, Iw, Tiv). The first has been already explained. The 
other two still occur in French loi, roi = (le^^a, ihwq)^ confused with 
(lua, rua) on the one hand and (Iwa, rwa) on the other, supra 
p. 187. The action is however truly simultaneous. The ags. 
wlaco (l«^aa"ko) seems to have generated (luuk) in lukewarm, and 

back of the tongue is raised, but the 
outer edges of the Hps are brought 
more together than for (u), and the 
sound of (w) when continued is there- 
fore a buzz, a mixture of voice and 
whisper, and not a pure vowel sound. 
When the buzz is strong the tremor of 
the lips is very perceptible, and a little 
more force produces the labial trill 
(brh). If the voice is removed we 
have (wh), and the back of the tongue 
being raised as before mentioned, the 
slightest effort suffices to raise it higher 
and produce (k^^;h). This gives the 
relation between the gutturals and 
labials which plays such an important 
part in comparative philology. On 
the other hand, for (bh) the tongue is 
not raised, the sound is a pure labial, 
less like (u), but easily deduced from 
(w) by lowering the tongue and slightly 
flattening the lips. It is, to those used 
to ' it, an extremely easy and pleasant 
consonant, produced with the least pos- 
sible effort. By dropping the voice it 
produces (ph), which is not now used 
in Europe, but was probably a value of 
<p. For (w, bh) there must be no contact 
with the teeth. Directly the lower 
lip touches the upper teeth, an impe- 
diment is raised to the passage of the 
air through the mouth, and the breath, 
escaping out on both sides, produces a 
rushing, rubbing, rustling sound, dis- 
tinctive of the "divided" consonants, 

and known as (v), which, on dropping 
the voice, becomes (f). But all degrees 
of contact between the lower lip and 
the teeth are possible, producing varie- 
ties of (f, v), from sounds which can 
scarcely be distinguished from (ph, bh), 
up to extremely harsh hisses and 
buzzes. Generally, then, (^w) is a con- 
sonant framed from (u) by closing the 
lips too closely to allow of a pure re- 
sonance for the vowel sound ; (bh) is a 
(b) with the lips just slightly opened, or 
a (v) without touching the teeth, that 
is, a pure labial ; (v) is a denti-labial. 
The (w) is further distinguished from 
(bh, v) by having the tongue raised. 
It is possible, of course to raise the 
tongue when sounding (v) ; the result is 
(vh), a very peculiar and disagreeable 
sound. But if the tongue is raised 
when sounding (bh) no ear would dis- 
tinguish the result from (w). The 
following words should be carefully 
pronounced to shew these differences : 
Fr. oui, oui; Eng. we, German wie, 
Fr. me = {vi,i lii wii bhii vii), Dutch 
letters u, v, w = (yy, vee, bhee) ; usual 
Scotch qulien, English when, Aber- 
deenshire fen = {kwh-ETi, when, fEn) ; 
usual German schreiben, faulty German 
schreiwen = (shrai-bm, shrai'bhm) ; 
German pferd now (pfert), once pro- 
bably (pphert) and in some Bavarian 
dialects (pnert). 


wlite (Ywirio) has become (loo'te), lote, cotuitenance in G. and E. 
1162, 2328. On the other hand, as wrong exists as (vraq) in 
Aberdeenshire, so wlcenco (Iw^qk'o) generated the Scotch wlonk 
(vloqk) the origin of our flunkey. In ags. wlips (l?^?ps) the labial 
modification has been simply dropped in Chaucer's lipsen 266, Sir 
T. Smith's (l«ps) and our lisp. Ags. v)l(Btian to nauseate, loath, 
seems to be lost, but (l^^at) and {laad^) = ags. la^, loath, are 
closely related in sound. JFl, wr, could scarcely be pronounced 
initially as (wl-, wr-), but would require the insertion of ('), thus 
(w'l-, w'r-), as seems to be the case in some Scotch dialects at the 
present day (p. 290.) The mode of writing would then be similar 
to that adopted for hi, hr ^= (Ih, rh). The reason why cw was 
used in preference to wc, is probably to be sought in the Latin qu, 
and the probability that (kw-) being sounded with tolerable ease 
may have been confused with the correct sound (k/^), for which 
there was a single character both in the E-unic and Gothic alphabets. 

The letter (g) of the Roman alphabet was also not quite the same 
as the ags. y in all cases. In later stages of the language, as in the 
xmth century, two forms (g, j) are found in use, the latter of 
which, under the form j became confused with z in writing, and 
subsequently in printing (p. 310). But the Roman ^ represented 
some of the sounds of ags. g and hence the Anglosaxons found no 
more difficulty in using it than is now felt by the modern high 
Germans. The two sounds (th, dh) however, had no Latin equiva- 
lent. Though the old Latins had introduced th, ch, for the Greek 
sounds d, %, the probability is that these letters were never properly 
pronounced, and that at the period in question they were merely (t, k) 
as at present in Italy, and therefore quite unsuited for Anglosaxon. 
Hence the necessity for \ ^, the former a rune, the latter a modified 
d, whereas the use of y for (y) would imply that the Latins still 
made some distinction between i and y. 

What were the precise meanings of j? ^, or rather how the mean- 
ings (th, dh) were distributed over them, it does not seem possible 
to elicit from the confused state of existing manuscripts. It is 
generally accepted that j? is (th) and ^ is (dh),^ yet \ is generally 
employed in initials, and ^ elsewhere, quite disregardful of modem 
usage, which we know has remained unaltered for 300 years, and 
therefore might be supposed to represent the old practice. We find, 
however, in modern Icelandic, a systematic adherence to the rule 

1 Mr. Oswald Cockayne seems to haps : wi'S, which some still call (weth), 
consider "8 = (th), and ]) = (dh), for in but then we also find: "Sough. These 
the preface to his edition of Hali inconsistencies in a modern writer who 
Meidenhad (supra p. 506, n. 1), which was evidently desirous of indicating the 
is otherwise in ordinary orthography, two sounds (th, dh) by appropriate 
he generally, but not quite consistently, letters may serve to explain the nu- 
employs ]? "S in these senses. Thus I merous inconsistencies of ancient and 
find : ]7e, j^is, }?ose, J^at, l>ey, j^em, perhaps less careful scribes, who were 
Jjeir, but : j^irteen]?, faij?, and in one certainly not less intending to carry 
place : au]7or, though in three other out theoretical conceptions of ortho- 
cases : au^or, is writen ; with this last graphy. See infra, No. 2, under "S j? 
spelling agrees : leng'S, dea'S, and, per- in the Icelandic Alphabet. 



Chap. V. 

of initial ]> and medial and final ^ in writing, and a uniform cor- 
responding pronunciation of (th) for J? and (dli) for 'S. Hence we 
should not be justified in pronouncing pure Anglosaxon in any 
other way, and we must suppose the change to have occurred^ in 
the transition period from pure Anglosaxon to Early English. 

In the above remarks we have endeavoured to assign the probable 
values of the Anglosaxon letters from the conclusions to which we 
were gradually led for the xiii th century, but these values differ 
materially from those assigned by our native Anglosaxon scholars. 
We have seen (p. 255, note 1) that one of them, an excellent 
scholar, who has paid much attention to the subject, decidedly calls 
long i (oi), long e (ii), long a (oo), long u (au). The well known 
scholar, Benjamin Thoipe, evidently made long i (ai), and short u 
(o), although he makes long e and u in Orrmin (ee, uu), see p. 487, 
note. Now it is certainly desirable to have some direct evidence as 
to the sounds of these long vowels, and this seems to be furnished 
by a valuable and interesting MS. at Oxford, to which attention 
was drawn by Hickes,'^ who gave some extracts from it, which will 
be here reproduced. In order to correct the errors in Hickes's 
transcription, Mr. G. Waring, of Oxford, obligingly collated the 
text with the MS., and has subsequently compared the proofs of 
the extracts with the original. I am also indebted to him for the 
account of the MS. given below.^ 

^ Usage is not yet quite fixed in some 
few cases. Meath and Lowth are com- 
monly called (Miith, Lautli) by the 
uninitiated, and (Miidh, Laudh) comes 
on them as a surprise. With the pre- 
position was always (w«th) in the 
XVI th century, and with the sub- 
stantive is still so called. Sometimes 
an arbitrary distinction is made. Dr. 
E. G. Latham calls himself (L^^th-em), 
but informs me that his family says 
(Le^dh"Bm). This is an instance of a 
variation of the medial th, which, so 
far as I can recall, is always (dh) in 
ordinary words. The change of final 
(dh) to (th) is natural enough, through 
the frequent use of (-dhth) as in breathe 
= (hriidhth) at the end of a sentence, 
or when prolonged without a following 
vowel. The initial change has only 
afi"ected the common words : that, the, 
thee, their, them, then, thence, there 
and its compounds, these, they, thine, 
this, those, thou, though, thus, thy. 
These have all (th) so far as they exist 
in Icelandic. But it must be remem- 
bered that we have a western dialect 
which uses (dh) initially in all cases. 
It would he interesting to know if 
there are any dialects which use (th) 
initially in all. Enclitically and after 
words ending with d, t we know that 

so late as Orrmin, and even later, ]? be- 
came t, and not d, even in Jiat, }7U, etc., 
and even after d, which is rather in 
favour' of a (th) than a (dh) sound. But 
see a different use, p. 444, note 2. 

2 Linguarum Vett. Septentriona- 
lium Thesaurus grammatico-criticus et 
archaeologicus. Auctore Georgia Hicke- 
sio, S.T.P. Oxford, 1705, folio, 3 vols; 
preface p. xi. 

2 The MS. is thus described by 
Hickes: "Dum in Bibliotheca Bod- 
leyana Codd. Saxonicos perscrutarer, 
inveni pervetustum librum MS. cujus 
nota, NE. D. 2. 19. iu quo quidem 
libro nonnulte lectiones e veteris tes- 
tamenti LXXII. interpretum versione 
Grseca, cum Latina translatione ex ad- 
verso in altera columna scripta, Saxo- 
nicis literis describuntur." Mr. War- 
ing says that the present signature of 
the MS. is Auct. F. 4. 32. It is a 
small quarto volume containing several 
unconnected pieces of great age and 
value. On the first page is a figure of 
Christ with an entry stating it to have 
been drawn by the hand of St. Dunstan. 
Fo. 1-8, " In honomatis sumi tonantis 
ars Euticis Gramatici," with several 
interlinear glosses, partly Latin and 
partly Old British.— Fo. 10-18. Anglo- 
Saxon homily on the Invention of the 

§ 4, No. 1. 



The peculiarity of this manuscript is that it gives certain Greek 
texts in Anglosaxon characters, which are seen immediately not to 
reproduce the original letters, but to be intended to represent the 
sounds in reading. There is no indication of the age of the MS. in 
any part of the book, but Mr. Waring thinks that these transcrip- 
tions were probably written in the latter half of the xth century.^ 
iN'ow we shall see that Greek was at that time probably pronounced 
almost, if not quite, as at present. Hence, by comparing the letters 
by which the Anglosaxon scribe translated the Greek sounds, we 
have direct evidence of the values he assigned to the Anglosaxon 
letters themselves. To make this comparison the more complete, I 
append the extracts given in Hickes, which are quite sufficient for 
the purpose, as collated by Mr. Waring, and contrast them with 
the modern Greek pronunciation, as obligingly furnished to me by 
Prof. Valetta,^ adding the ancient text for comparison.^ As the 

Cross, superscribed Ixiii, as if forming 
part of a collection. The handwriting 
is ancient, the language pure and 
strictly grammatical. Judging from 
these characteristics and certain pecu- 
liarities of dialect, Mr. Waring assigns 
it to the latter half of the x th centui-y. 
The legend is that of the poem of 
Elene.— Fo. 19. See below at fo. 24.— 
Fo. 20-22. A Lunar and Paschal Ca- 
lendar. — Fo. 23. Pauca de Mensuris, 
containing several Old British glosses. 
— Fo. 19 and fo. 24-36. Extracts from 
the Septuagint with corresponding 
texts from the Itala, in two parts : fo. 
24-28, the Septuagint text in Greek 
characters, full of flagrant blunders, 
and critically worthless; fo. 19, and 
half of fo. 28 to 36, the Septuagint 
text in Anglosaxon characters, of a 
decidedly better quality than the other. 
— Fo. 37 to end, Ovidii Nasonis Artis 
Amatoriae, Lib. prim., accompanied 
with many interlinear glosses in Latin 
and Old British. — The pieces com- 
mencing on folios 1, 20, 23, 37, are 
noticed in Lhuyd, Archseol. p 226, and 
Zeuss, Celtica I, p. xxxviii, and II, p. 
1076 ff. The whole codex is described 
in p. 63 of : Antiquae Literaturse Sep- 
tentrionalis Liber Alter seu Hmnphredi 
Wanleii Librorum Yett. Septentrion- 
alium, qui in Anglise Bibliothecis ex- 
tant, nee non multorum Vett. Codd. 
Septentrionalium alibi extantium Cata- 
logus Historico-Criticus, cum totius 
Thesauri Linguarum Septentrionalium 
sex Indicibus, Oxford, 1703, folio, 
forming the second volume of Hickes. 
The Scribe apparently did not know 
Greek. The letters are run much into 

each other, with very imperfect at~ 
tempts at arrangement into words. 

^ The following are his reasons : 
There could be little doubt of the date, 
if a period could be assigned when 
priests of the Anglosaxon church might 
have been brought into connection with 
those of Constantinople, and this is 
easily done. Otho I, emperor of Ger- 
many, 936 - 973, married Eadgith, 
daughter of King Athelstan I of 
England, 930, and his son and suc- 
cessor Otho II, married Theophania, 
daughter of the Greek Emperor Nice- 
phorus, in 972, after the latter em- 
peror's assassination. At the court of 
Otho, then, where constant connection 
was kept up with the Anglosaxons and 
the Greeks, there was a means opened 
out for the priests of the former to 
receive some tincture of Hellenic 
letters. "We shall therefore hardly be 
wrong in referring such transcriptions 
to the latter part of the xth century. 
Want of opportunity is against an 
earlier date, and the confusion and ruin 
occasioned by the Danish invasion in 
the early part of the xi th century, the 
close connection of Canute with Rome, 
and the subsequent Norman influence 
through Edward the Confessor, render 
a later date almost impossible. To this 
we may add the agreement of the Saxon 
homily in the same book with the 
language of the x th century. 

2 Author of a learned work in mo- 
dern Greek on the Life and Poems of 
Homer. 'Ofirjpov /Sios koI 7rofJ7/xaTa, 
irpayixaTeia laropiKi} /cat KpniK^y xmh 
'Iwdvvov N. BaAeVra, London, 1867. 

^ There will be found many dif- 



Chap. V. 

modem Greek does not distinguish long and short vowels, and does 
not seem to appreciate any such difference, but pronounces the same 
vowel in the same word sometimes long and sometimes short, ac- 
cording to the feeling of the moment, I felt that it would be mis- 
leading to indicate long and short vowels in the following, and I 
have therefore, for convenience marked them all as short. The 
same indistinctness exists in the Italian, Spanish,^ and French 
languages, and probably exists naturally wherever the vowels are 
in perfect pairs. On a very accurate examination of the vowel 
pairs in English it will be seen that in many words they differ 
rather in quality than in quantity, and that there is, as Professor 
Haldeman urges, a medial length of vowel, ^ which is sometimes 
heard as short and sometimes as long. The Scotch consider most of 
their vowels as short, though they strike an English ear at first as 
long, being probably medial, and Eeline marks almost all Erench 
vowels as short, though other writers mark them frequently as long. 
When I have placed the accent mark after the vowel instead of 
after the consonant, there seemed to be certainly an option in pro- 
nouncing long or siaort, and the shortest vowels, are, as in Italian, 
always perfectly clear and never degenerate into obscuiities like the 
English. The letters jB, cf), seem to be naturally pronounced by 
Prof. Valetta as (bh, ph), but when he became particularly em- 
phatic he made them (v, f ). I have, therefore, used (v, f ) in my 
transcription as more convenient,^ and for the same reason have 
transcribed av, ev as (av, ev) or (af, ef ). 

ferences between the two editions, but 
it was thought best to follow the usual 
text of the Septuagint. 

1 My attention was first drawn to 
the doubtful medial quantity of the 
Italian vowels by H.I.H. Prince Louis 
Lucien Bonaparte, and Senor Cubi y 
Soler made me notice the absence of 
truly ' stopped,' or shut, short vowels 
in Castilian, which he said was a par- 
ticular mark of that leading Spanish 
dialect, so that he suggested the use of 
long vowel signs in all Spanish words. 

2 Analytic Orthography, p. 80. Prof. 
Haldeman makes short vowels last 
about a quarter of a second, medial 
about three-eighths or one-half, and 
long vowels about five-eighths or three- 
quarters, so that the comparative 
lengths are about as 2, 3 and 5, or 
1, 2, 3. 

s The sounds (bh, ph) are most pro- 
bably very ancient. Prof. Goldstiicker 
in his learned paper on the Greek Di- 
gamma, read before the Philological 
Society, 20th Nov. 1868, attempted to 
point out the Greek words in which it 
had existed by means of a comparison 

with the Sanskrit form, inferring a 
digamma in many cases where the 
latter began with (v), or (sv), and the 
Greek had either no initial consonant 
or only an aspirate. Eemembering 
that the Sanskrit grammarians afiirm 
the Sanskrit sound to be a true (v), 
made with action of the teeth, and 
that in Spanish we know historically 
that Latin F, certainly (f), passed 
through (h) and became lost, as in 
Jilius, old Spanish Jijo, modern hijo in 
which the h is not pronounced (i'xo), 
and knowing first how easily (v, f ) are 
confused, next how unlikely the Greeks 
who had acp = (sph) would be to allow 
(sv) or (sf), the ease with which there- 
fore an initial (s) in this combination 
would be rejected, and at the same 
time the very probable transit of (s) 
into (f), we are led to the sound of (f) 
as that most likely to fulfil the phone- 
tic conditions imposed on the digamma 
by comparative philology. The sound 
(w) would not be easily lost except 
before (o, u), and the sound (bh) was 
already probably existent, and became 
fixed as one (if not the only) sound of 

$ 4, No. 1. 



Extracts teom the Bodl. MS. Atjct. F. 4, 32. 

Anglosaxon Transcription. 
MS. fo. 30, i. 

26. Phyisomen anthropon cat 
icona ce cath omyosin imeteran 
ce archeto ton icthyon tis talasas 
ce ton petinon tu uranu ce ton 
ctinon ce passes tis gis ce panton 
ton herpeton ton herponton epi 
tis gis ce egeneto utos 

27. ce ephyisen o theos ton 
anthropon cat icona then epyisen 
auton aren ce thily epyoeisen 

28. ce eulogisen autus legon 
auxanesthe ce plithynesthe ce 
plirosate tin gin ce catacyrieu- 
sate autis ce archete ton icthyon 
tis thalassis ce ton petinon tu 
uranu ce ton panton ctinon tis 
gis ce panton ton erpeton ton 
erponton epi tis gis 

29. ce ipen o theos idu edoca 
ymin panta chorton spomonri 
spiron sperma 6 estin epano pas- 
sis tis gis ce pan xylon o cchi 
en eauto carpon spermatos spori- 
mu ymin estae is hrosin. 

30. ce passin tys thiriys tis 
gis ce pasin tys petinys tu uranu 
ce panti erpeto erponti epi tis 
gis 6 echi en eauto pnoin zois ce 
panta chorton chloron is brosin 
ce egeneto utos. 

Greek Text. 

26. irotrjcajjUer &vQpu)iTov Kar* cIkSvu 
rifierepau koI Kad' ofiolwaiv /cat apx^- 
TOiffav tS)V IxOvcay rrjs Qaka.(ra"r\s, Koi 
Tobv irer^LVccu tov ovpauov, Kai tuv 
KTTjvwv, Kal ird(rr]S rrjs yrjs, Kol irdvrcov 
Twv epireruv tS)v kpirovTuv 67rl t^s 77)5. 

27. KoX eTToiTjcrej/ 6 ©eby rhu &v6pci)- 
TTOV Kar^ cIkSvu Qeou iiroi'qa'^v avTSv 
dpaev Kol OrjXv iTroiriffei/ avrovs. 

28. Koi ^v\6y7](X^v avrovs 6 0ebs, 
Xeywu, av^dveaOe, koI Tr\r}6vu€(r0€, koL 
irArjpdoaaTe tt;!/ yrjv, Kai KaraKvpL€v<raT€ 
avTYjS' Kal ^pxere tcov IxOvooj/ rrjs 
6a\da'a7}Sf Kal rSiv Trereivuv rov ovpa- 

Modern Greek Pronunciation. 
Genesis ch. i. 

26. Piji'somen an'thropon kat 
iko'na imeter'an ke kath omi*- 
osin, ke arkhet'osan ton ikhthi'- 
on tis thala'sis, ke ton petinon- 
tu uranu*, ke ton ktinon*, ke 
pa "sis tis jis, ke pan "don ton 
erpeton* ton erpon*don epi* tis jis. 

27. ke epi*isen theos* ton 
an*thropon. kat iko*na the,u* 
epi*isen afton*, ar*sen ke thi*li 
epi*isen aftus* 

28. ke evlo'jisen aftus* 
theos*, legh'on, afksa'nesthe, ke 
plithi'nesthe, ke pliro*sate tin 
jin, ke katakirief*sate aftis*, ke 
ar'khete ton ikhthi'on tis thala*- 
sis, ke ton petinon* tu uranu*, 
ke pan*don ton ktinon*, ke pa*sis 
tis jis, ke pan*don ton erpeton* 
ton erpon*don epi* tis jis. 

29. ke i*pcn theos*, idhu* 
dhe'dhoka imin* pan* da khor*ton 
spo'rimon spi'ron sper*ma, es*- 
tin epa'no pa* sis tis jis, ke pan 
ksi*lon, e*khi en eafto* karpon* 
sper'matos spori*niu, imin* es*te 
is vro*sin* 

30. ke pa*si tis thiri*is tis jis, 
ke pa*si tis petinis* tu uranu*, 
ke pandi* erpeto* er*pondi epi* 
tis jis, o ekh'i en eafto* psikhin* 
zOjis*, ke pan*da khor*ton khlo- 
ron* is vro*sin, ke ejen*eto u*tos. 

Genesis, Ch. i. 

vov^ KoX nrduTOiV roov kttjj/wj/, koX TrdffTjs 
T7)s yrjs, Kal iravruiv rai]/ epTreTcoy ruv 
epirovTuv iirl rrjs yr^s. 

29. Kal elirey 6 idehs, iSov Se'SwKa 
vfuv irdvTa x^P''''^^ (nrSpifiov aireTpov 
(TTrepfxa, iariv etrdvca Tvdcrrjs ttjs yrjS' 
Kal irav ^v\oy, t 6%^' ^'^ iavTw KapitSv 
(TTTepfiaTOs (TTropifiOv, viuv ecrrai els 

30. Kal iraffi to7s Orjpiois Trjs yrjs 
Kal rracri rots ireTeivoTs rov ovpavov, 
Kal iravrl epirercf} epirouTi iirl ttjs yrjs, 
t> exet iy eavT^ ypvx'h'^ C^V^i '<^«^ rrdyra 
X^pTou xAcopb;/ ets fipHeiy, Kal iyevero 



Chap. V. 

Anglosaxon Transcription. 

31. ce yden o thcos ta panta 
osa cphyisen ce idu cala lian cc 
egeneto hespera ce egeneto prohi 
himera ecti . 

MS. fo. 34, b. 

1 . theiis epirasen ton habra- 
cham ce ipen pros auton habra- 
cham habracham ce ipen idu ego. 

2. ce ipen labeto yion su ton 
agapeton on egapesas ton isac ce 
poreutheti is ten gen ten ypselen 
ce prosenencon auton eci is olo- 
carposin is ena ores on sy ipo 

3. anastas de habracbam to 

MS. fo. 34 a. 

1. on tropon epipothie elafos 
epi tas pegas ton ydaton utos 
epipotbi e psuYcbe mu pros se o 

2. edipsisen e psycbe mu pros 
ton tbeon ton zonta pote ixo cae 
optbesome tu prosopu tu tbeu 

3. egenetbe my ta dacrya mu 
artos emeras cae nyctos. 

MS. fo. 32, b. 
1 . Ce epilempsonte epta gyne- 
ces enos antbropu leguse ton 
arton emon fagometba ce ta 

Modern Greek Pronunciation. 

31. ke i'dhen o tbeos* ta pan*- 
da, os'a epi'ise, ke idhu* kala* 
li'an* ke ejcn'eto esper'a, ke 
ejen'eto pro,i', imer'a ek'ti. 

Genesis eh. xxii. 

1. tbeos* epi'rasen ton Avra,- 
am", ke i'pen afto', Avra,am*, 
Avra,am', ke i'pen idbu" egho* 

2. Ke i"pe, lave ton i,on' su 
ton aghapiton*, on igba'pisas ton 
Isa,ak*, ke porcf'thiti is tin jin 
tin ipsilin*, ke anen'cqke afton* 
eki* is olokar'posin ef en ton 
ore'oon on an si i*po. 

3. anastas • dbe Avra,am' to 
pro,i- .... 

Psalm xlii. 

1. on trop'on epipotbi* i el*- 
afos epi" tas pigbas* ton idba'ton 
u'tos epipotbi* i psikbi* mu pros 
se, tbeos* 

2. edbip'siseni psikbi* mu pros 
ton tbeon* ton zon*da; po*te 
iks'o ke oftbi'some to proso*po 
tu tbe,u* ? 

3. ejeni-tbi ta dbak*ria*mu 
emi* ar'tos imer*as ke niktos* 

Isaiah ch. iv, 
1. ke epilip* sonde epta* jine*- 
kes antbro'pu enos*, legb'use : 
ton ar*ton imon* fagbom*etba, ke 

Greek Text. 

31. kclI eTSer o 0ebs rh. irdvTa, '6(Ta 
hrolrjcre' Koi ISoh Ka\a Xiav' koX iyeueTO 
lenrepa, koI iyevero irpuf rjfiepa €KT7J. 

Genesis Cb. xxii. 

1. 6 Qehs itr^lpaff^v rou 'A^paa/x, 
Kol elirev avrw' 'A)8/3aa/u, 'Afipadfjt.' 
Kol elirey, idov 4ycv. 

2. Kal efire, \d$e rhv vi6v ffov rhv 
ayoTTTfrhv, hu i}yd7rr](ras, rhv 'Itroa/c, Kal 
iropevd'qTi els ttjv y-qv rrfv v\pr]\})y, Kal 
areVey/ce avrov e/ce? us oKoKapiruatv 
i<p' ev rS)V op4(av uv av aoi eJfTrw. 

3. avaffTas Se 'A^paafi rh irpuit .... 

Psalm xlii. 

1. '6v Tp6Trop eiriiroOu rj e\a(pos 4irl 
ras Trrjyas rHov uSdruv, ovTuSy iirLirodei 
V ^'"xh H-ov irphs (re, 6 QcSs' 

2. iSixprjCev rj ^vx'f] jJ-ov irphs rhv 
&ehv rov ^wvTa' irore 7^|&> Kal ocpd-fjCofjiai 
T(j) TTpoawiro} Tov &€ov ; 

3* iyei/rjOrj ra SdKpvd fxov ifiol dpros 
r]iJ.epas Kal vvktos. 

Isaiah Chap. iv. 
1. Kal iinK^i\iovrai eirra yvva^Kes 
hvOpctiirov €j/hsy Xiyovaai' rhv &prov 
Tjfiuv (I)ay6fxeday Kal ra ifidria rj/iwv 

§ 4, No. 1. 



Anglosaxon Transcription. 
imatia emon peribalometha plen 
to onoma su ce elite ef emas 
afele ton onidismon emon 

2. te de emera ecinie empi- 
lampsi theus en boile meta 
doxes epi tes ges tu ypsose ce 
doxase to cataliptlien tu israhel. 

3. ce este to ypolipthen en 
sion ce to catalipthen en hiru- 
salem agiy clethesonte pantes y 
engraphentes is zoen en hirusa- 

4. oti ecplyni kirios ton rnpon 
ton yion ce thygateren sion ce 
to aema eccathari ec messo auton 
en pneumati criseos ce en pneu- 
mati causeos. 

5. ce exi ce este apas topos 
tu orus sion ce panta ta peri 
cycle antes sciasi nefele emeras 
cae OS capnu ce fotos pyros 
ceomenu nyctos pase te doxe 

6. ce este is scian apo cau- 
matos en scope ce en apocryfo 
apo scelerotetos ce yetu. 

Isaiah ch. V. 

1. Aso de to agapameno asma 
to agapeto to ampeloni mu Am- 
pelos egeneto to ecapemeno en 
cerati en topo pioni 

TrepifiaXov/uLeOa' irX^jv rb ovofxa rh crhu 

2. T77 Se 7]ixepa iKeiurj e7rtAa/xi|/et 6 
©ehs iv fiovXrj fiera 56^r}S cttJ rrjs yrjs, 
Tov v^l/cccrai Koi So^dcrai to KaraXeicpdhi/ 
Tov 'IcrpaTjA. 

3. Kol iarai, rh viroAeicpO^v eV '^li^v, 
Koi rh Kara\ei(p9eu iu '\€pov(TaX)][x, 
ayioi KK7]0^(TOprai irdvres ot ypacpdures 
eU C^'hi' ^v 'lepovcraA-fifjL. 

4. '6ri iKwAvuel Kvptos rhu pvirov rdv 
viuv Ka\ rcov dvyarepcou 'Sioou, Koi rh 
aTixa (KKaOapieT iK /xeaou avrcov, iv 
vpevfMari Kpiaews Koi iri/evfj-ari KaiKrecos. 

Modern Greek Pronunciation. 
ta ima'tia imon* perivalu'metha: 
plin to o'noma to son keklis'tho 
ef imas*, af'ele ton onidhismon* 

2. ti dhe imer'a eki*ni epi- 
lam'psi o theos' en vuli* meta* 
dhok'sis epi" tis jis, tu ipso'se ke 
dhoksaa'se to katalifthen* tu 

3. ke es'te to ipolifthen* en 
Sion' ke to katalifthen* en leru- 
salim*, a'jiji klithi'sonde pan'des 
i ghrafen-des is zoin* en leru- 

4. o'ti ekplini* ki'rios ton 
ri'pon ton ion- ke ton thigha- 
ter'on Sion*, ke to e*ma ekkath- 
ariji" ek mes'u afton*, en pnev- 
mati kri'seos ke pnevmati kaf*- 

5. ke ik'si*, ke es'te pas to'pos 
tu or"us Sion", ke pan*da ta peri- 
kik'lo aftes" skia'si nefel'i imer*- 
as, ke OS kapnu* ke fotos* piros* 
keomen'u niktos*, ke pa*si ti 
dhok'si skepasthi'sete. 

6. ke es'te is skian* apo* kav- 
matos, ke en skep'i, ke en 
apokri*fo apo* sklirot'itos ke ietu*. 

Isaiah ch. v. 
1. a*so dhi to ighapimen*o 
as*ma tu aghapitu* mu to ambe- 
lo*ni mu. Ambelon* ejeni*tlii to 
igapimen*o eq ge'rati en do*po 

5. Koi i/j^ei, Kol iarai iras r6iTos rov 
opovs '2,i(i)V, KoX Trdvra ra TrepiKOKXcp 
avrrjs (TKidcrei vecpeXr] rj/xepas, Kal ws 
KUTTUOv Kal (pcarhs irvphs Kaiofxeuov vvK- 
rbs, KoX irdar} rfj SS^r/ ffK^iraaQiia'irai. 

6. KoX icTrai els aKidu airh Kavfiaros, 
Kal eV (TKeirr), Kal eV a.TTOKpv(p(^ airh 
cTKXripoTTjros koI verov. 

Isaiah Chap. v. 

1. acrco StJ ru) r] ac/xa rov 
ayaTrr]rov fxov r^ aixinXuivi ixcv. 'Afi- 
TreXdov iyevijOr) rc^ riyain]fj.4ucf, iy 
Kcpari, iv rSirci) iriout. 



Chap. V. 

Anglosaxon Transcription. 

2. ce fragmon periethcca cae 
echaracosa cc cphyteusa ompelon 
sorec ce ocodomesa pyrgon en 
meso autu ce prolenion oryxa 
en auto ce cmina tu pyese stafy- 
len epyesen de acantas 

MS. fo. 33, b. 

1. Y dipsontes poreuesthe ef 
ydor ce osy men u cecethe 
argyra°n badisantes agorasete ce 
piete aneu argyriu ca3 timis ynon 
ce stear 

2. inati timasthe argyrio ke 
ton misthon ymon .u. chi plis- 
moni'n acusate mu cae fagesthe 
ta agatha ce tryfisi en agathys 
i psychi ymon 

3. prosechete tys osin ymon 
ce epacoluthisate tes odys mu 
acussate mu cae ziste en agathys 
i psyclii ymon cae chathisome 
ymin diathicin eonion ta osia 
dauid ta pista. 

4. idu martyrion auton dedoca 
ethnesin arclionta ce prostas- 
sonta ethnesin. 

5. ethni a uc idisan se epicale- 
sonte se cae y las .y. uc epistanto 
se epi se catafeuxonte enecen tu 
then tu agiu israliel oti edoxasen 


2. Kol (ppayfxhv irepiedrjKa, Ka\ ix^poi- 
Kcoffa, Kal icpvrevaa &fxiT€\ou Swp^/c, 
Kol <fKod6fxr]aa irvpyov iu ixeacp avrov, 
Kal TTpo\T]viov iopv^a eV avrcf, Koi cfxeiva 
Tov TToirjaai ara^vA^u, Kal iirdnjaev 

Isaiah Chap. Iv. 

1. ol di\pcii}i'T€s, TTopeveaOe e<|)' uSwp, 
Kal o(roi /xTj ^x^Te apyvpLOj/, fiaSiaauTcs 
ayopdaare, Kal (pdy^re &yev apyvpiov 
Kal rifx^s oivov Kal (Treap. 

2. luarl TifxaaOe apyvpiov, Kal rhv 
fxSx^ou vfjLcov ovK els TrX'i)(rfxov'i]i' ; 
CLKovcrare jnov, Kal (pdyeffOe ayaOa, Kal 
iuTpvcp-fiaei iv ayaQois i] ^vxh v/u-wv. 

Modern Greek Pronunciation. 

2. ke fraghmon" perieth'ika 
ke ekhara'kosa ke efi'tefsa am'- 
belon Sorik' ke okodlwmisa pir'- 
ghon en mes'o aftu* ke proli'nion 
oTiksa en afto*, ke em'ina tu 
piji'se stafilin', ke epi'isen akan*- 

Isaiah ch. Iv. 

1. i dhipson'des, porevesthe 
ef i'dhor, ke o"si mi ekh'cte ar- 
ji'rion, vadhi'sandes aghora'sate, 
ke fa'jete an'ev arjiri'u ke timis* 
i'non ke ste'ar. 

2. inati" timas'the auiri'u, ke 
ton mokh'tlion imon* uk is plis- 
monin*? aku'sate' mu, ke fa*- 
jesthe aghatha', ke endrifi'si en 
aghathis- i psikhi* imon*. 

3. prose'khete tis osin* imon*, 
ke epakoluthi'sate tes odhis'mu : 
isaku'sate* mu ke zi'sete en 
aghathis* i psikhi* imon*, ke 
dhiathi'some imin* dhiathi'kin 
e,o'nion, ta o"sia Dhavidh' ta 

4. idhu* marti*rion en eth'ne- 
sin e'dhoka afton* ar'khonda ke 
prostas'onda eth*nesin. 

5. eth'ni a uuk i'dhasi* se 
epikale* sonde* se, ke la,i* i uk 
epis'tande* se epi* se katafef*- 
ksonde en*eken Kiri'u tu the,u* 
su tu aji*u Isra,il*, ot'i edhok*- 
sase* se. 


3. ■7rpoa'e%6T6 ro7s axrlu vfiwv, Kal 
i'KaKo\ovi)7\crare toas 6do7s fiov cis- 
aKovffare fiov, Kal (rjcreTai iv ayaQois 
7] ^vxv vjxuv, Kal hia6T]( vfuv 
Sia9r]Kr]v alwvioVy ra '6(ria AavlS to, 

4. iSob fxaprvpiov iv edvecriv eSwKa 
avTov, 'dpxovTa koX irpoffTdcro'ovTa 

5. eOvT} a OVK oWaffi (re, iiriKaXiffov 
ral ere, Kal Xaol oi ovk iirla'TavTai ae 
iirl (Te Karacpev^ovrai, eVe/cer Kvpiov 
TOV &eov ffov TOV ayiov 'lapa^\, '6ri 
iS6^a(r4 <re. 

§ 4, No. 1. 



Erom these extracts we may deduce tlie following table of tlie 
correspondence of the Greek and Anglosaxon letters. A third 
column shews the values now attributed to the Greek letters in 
Athens, including some combinations which do not occur in the 



^ a 

01 o 
<1> ■ r3 


rM fl 

0) o 




(D O 

O O 


O O 


to rj 
.2 " 




rrt fl 



-S fl 



-S c" 


C cj 

o o 


O ns 

o O 


c n 

o '^ 



5 1- 



















s ss 



e ae 









au av 

av af 
















g c 

gh J 








qg g 



mb b 


ph f 















nd d 
































eu ev 

ev ef 





ov of 





P Pb 






i e 














As Prof. Yaletta pronounced, a was (aa, a) or [aa, a), but there 
was never any rounding or labialisation producing (ah, a). Prom 
this, however, it does not follow that the ags. a which transcribes 
a may not have had a labialised form, for, just as the Prench a 
was called (a) in England, when it was only (a), p. 226, note, col. 
2, so the Angiosaxons would have transcribed a by a, even if the 
first said (a) and the last (a). But we may safely conclude that 
ags. long a was not (oo) or even (oo). 

The uniform transcription of e, and almost uniform transcription 
of at, by e, precludes the idea that ags. e was ever anything but 
(ee, e). When ac was not represented by e, which is very rarely, 
it is represented by ae, which must be regarded rather as a Latin 
than an ags. form, having then the invariable sound of (ee), 
although the ags. se itself is found in cee Is. 55, 1. 5. Thus koI 
is generally written ce but occasionally cae Ps. 42, 2 ; and este 
earai Is. 4, 3. 5 is evidently more correct than estae, Gen. 1, 29 ; 
so that aema alfia Is. 4, 4, should be ema. 

The transcription of ft) o by o, shews that ags. must have been 
(oo, o) or {oo, o). Prof. Valetta pronounced Greek, and indeed 
English, with a clear (oo, o), and did not seem to be aware of (oo). 
But just as Englishmen nowadays report the Greek co to be (oo). 
so the Angiosaxons would of course have used their o, whether it 
meant (oo) or {oo). 

The uniform transcription of c by i shews that ags. i was certainly 


(ii, i) or (n i). There are six letters and combinations in modem 
Greek which, in Prof. Valetta's pronunciation, have the sound of 
(ii, i), viz. : t) v v ei ol vl. Of these the ags. transcription gives * 
for c and eL uniformly, with the single error te in Ps. 42, 1 epipothie 
iTniroOel. For tj we find most generally e, but in about 50 in- 
stances e, not, however, uniformly, thus against passes Tracrr)^; Gen. 
1, 26, we ^nd passis, ib. v. 29 ; against ten gen ttjv 'yr]v Gen. 22, 2, 
we may put lis gis, Gen. 1, 30 ; against emeras rnjuepa^ Ps. 42, 3, 
we have himera Gen. 1, 31 ; against psyche '>^V')(r} Ps. 42, 2, we 
have psychij Is. 55, 2 ; against epyesen iirolrjaev Is. 5, 2, we have 
epyisen Gen. 1, 27, against exi rj^ei Is. 4, 5, we have ixo ri^(o 
Ps. 42, 2, and so on. Hence we cannot conclude that rj was 
sounded as (e), or e as (i), but must consider that there was some 
confusion in the mind of the scribe, perhaps arising from the Latin 
transcriptions of rj, with which he was necessarily more familiar. 
The forms ecinie i/ceivy Is. 4, 2, and agapameno rjyaTrrj/jLevw Is. 5, 1 
are mere mistakes. The Greek v oi are uniformly rendered by y 
and VL by yi, mere clerical errors excepted, as epyoeisen iiro'irjo'ev 
Gen. 1, 27 when five words before it was epyisen ; and ecpluni rupon 
eKirXwel pvirov Is. 4, 4, between which words stands hrios 
Kvpto^ (having i and not y for f,) as if to shew the error, while 
psuVche "^^X^ -P®- '^^j -^j indicates an intention to correct such 
errors. Now we have reason to suppose that the earlier sounds of 
V VL OL were (y, yi, ui), and that the degi^adation of y, yi into (i), 
was similar to the common upper German use of (i) for (y), while 
(i) for (ui) is comparable to the ^renoh. frangais (fraASE) for frangois 
(fraAsuE). At present Prof. Yaletta will not admit any other sound 
but (i) for any one of the three combinations, v vl ol, but Pranz 
asserts in his Modern Greek Grammar,^ that v vl ol resemble Prench 
w,^ which at least shews a probability that the Anglosaxon scribe 
also recognized (y) rather than (i) in the combinations v vl ot, and 
hence that the ags. y was, as is generally suspected, (y). 

The Greek ov is the least disputed of the Greek sounds ; it re- 
mained for writers of the xvi th century to start the theory that 
both Greek ov and Latin u were (ou), supra pp. 150-1. We find 
it uniformly represented by u, with the exception of the manifest 
error hoile ^ov\y Is. 4, 2. 

As to the transcriptions au, eu for au, ev, it is not easy to say 
whether they are to be taken as Latin (au, eu), or whether u is 

^ Grammatica Linguae Grsecae Ee- minanti imprimis hse tres 77 t v sese 
centioris, Romae in Collegio Urbano, offerunt, de quibus si quis ex usu vul- 
1837, 8vo. pp. V, 137, and tables. The gari judicaverit, facile adduci potest, ut 
preface is signed Joannes Franzius, and nullum in sono earum discrimen de- 
dated Eomae, Idibus Martiis, 1837. prehendi arbitretur. Quanquam illud 
Franz was, I believe, a Bavarian priest quidem negari non potest, quum tj 
who was sometimes at the court of magis ex imo pronuntietur, v ad sonum 
Otho. Gailici u propius accedere .... ot ut 

2 "Vocalium pronuntiationem exa- u (gall.)7roios, ut^s (pyos, yos)." Ib.p.2. 


''u consonant," that is v, in which case (av, ev) would agree with 
the modern sounds except before tt, t, k. 

These transcriptions establish, therefore, by direct evidence, that : 
ags. a was one of the sounds (a, a, ah, a), and not (o, o). 
ags. e was (e). 

ags. ^ was one of the sounds (i, i), and not a diphthong like (ai) 
ags. was one of the sounds (o, o) 
ags. u was one of the sounds (u, u), and not (ou) 
ags. y was probably (y) but may have been (i) or {i) 
The transcription has several foreign letters and combinations as, 
ae, z, th, X, ph, ch, the meaning of which is generally evident. The 
only difficulty is ph when used for tt in phyisomen Trotrjcrcojuiev, Gen. 
1, 26, ephyisen eTroirjaev, v. 27, where it is explained by the con- 
current form epyisen in the same verse. In all other words p only 
is used. The concurrent form / when ph represents (f) as in nefele 
fotos vecpeXr) <^coto9, Is. 4, 5, shews its value in this case. Before 
th, there seems to have been the same difficulty of pronouncing phy 
as at the present day, where so many say, as most used to write 
dipthong (d^'p'thoq), for we find opthesome ocpOijaofiac Ps. 42, 2, 
ypolipthen V7ro\€i<pOev Is. 4, 3, where the modern Greek says 
(ipolifthcn*). Similarly cth is used for ^^ in icthyon l')(6vcov Gen. 
1, 28. It is rather remarkable that J? was not used for 0. 

The consistent use of c to transcribe Greek /c, to the exclusion of 
h, shews that the ags. always pronounced c as cither (k) or (Ic), the 
distinction, of course, being unrecognized. As b, g, d are used for 
/?, 7, h, no countenance is given to the modem uses (bh, gh, dh), 
where (bh) becomes (v), and (gh) is rather (grh) or the lighter (r), 
but before (i, e) falls into {gh., grh) or (j), the last being the re- 
cognized sound. The character ^ stood in readiness for 3, but as 
th had been used for Q, dh would have been the only appropriate 
sign for h, and this was not a known symbol. Perhaps the use of 
]>, ^, had begun to be unsettled, and this may have prevented their 
employment for 0, S. The ags. g was itself most probably often 
(gh) and hence no better sign could be devised, even if the (gh) 
sound of 7 was recognized. The modem change of tt, t, k, into 
(b, d, g), after fi, v, 7, is not acknowledged. But the change of 
7 into (q) before k in the middle of a word is acknowledged as 
prosenencon avkve^Ke Gen. 22, 2. 

The Greek aspirate is generally omitted, but an h is occasionally 
inserted where there is none in the original, especial to avoid an 
hiatus as prohi irpcot, Gen. 1, 31, israhel 'IcrparjX, Is. 55, 5, and 
this is occasionally strengthened in ch as halracham ^A^adfjb. 

The principal gain, then, of this transliteration is the establish- 
ment of the Angiosaxon simple vowel system within certain limits ; 
nothing is gained for the double vowels ea, eo. On the whole, the 
results are confirmatory of those arrived at by the totally difi'erent 
process of gradual ascension from the English of the xiv th, xiii th, 
and XII th centuries. 


Wc have assumed as well known that the pronunciation of Greek 
in the x th century at Byzantium was practically the same as that 
now in use at Athens.^ The proofs of this are to be sought in the 
hieroglyphical transcription of the names and titles of the Greek 
and lloman Pharaohs, as collected in Lepsius's Konigsbuch, in 
the Septuagint and the I^ew Testament transcription of Hebrew 
words, and in the New Testament transcription of Latin names, in the 
Syriac vowel points, in the transcription of Latin names by Polybius 
and other Greek writers, in the numerous errors of the old Christian 
and other inscriptions, and, among other sources, in the writing of 
Latin words in Greek letters in the vi th and vn th centuries, by 
certain Greeks at liavenna, who had to attest certain Latin documents 
which still exist, and have been published by Marini.'^ As a com- 
panion to the above transcription of Greek into Anglosaxon characters, 
a few of these attempts by Greeks to write Latin in Greek characters 
will be interesting, and, if we bear in mind that they were writing an 
unknown language from dictation and would be therefore likely to 
commit as many errors of audition and pronunciation as a decidedly 
provincial Frenchman, ignorant of English, who attempted to write 
English from dictation in his own characters, we shall see that the 
key to his meaning is to be found in the modem pronunciation of 
Greek. The Latin interpretation here annexed has been deduced 
from corresponding Latin attestations in the same documents. The 
Latin letters w, n, d, indicate some peculiar fonns of v, v, B, and h 
is sometimes Latin h, and sometimes a peculiar form of ij. The 
transcript of Marini is not always trustworthy, and in a few 

1 ""Why Greek alters not in fourteen vincial speakers among the highest of 

centuries, and English must needs alter the realm, the general importance of 

in four, is queer," wrote a friend in secondary cities, and other causes, 

reply to an observation of mine on the readily suggest themselves to account 

pronunciation of Greek at the time of for the numerous changes which have 

Ulfilas, Of course there must have prevailed. If we examined the Greek 

been reasons for the preservation of dialects at present for variety of pro- 

any pronunciation for so long a time. nunciation, we should probably obtain 

Greece was a very small country, but a large amount of information, impor- 

it had numerous dialects, and by ne- tant in its bearings even upon ancient 

glecting these we reduce the country Greek usages. The modern system of 

almost to one city, Byzantium, the seat education however, which aims at uni- 

of the Greek empire, and of Greek formity of pronunciation and a recur- 

learnmg and literature, till quite recent rence to ancient idiom, only the ancient 

times. The pronunciation we have to Greek Grammar being taught in schools, 

deal with is therefore that of an undis- may soon efface these records of the 

turbed court and literary dialect, in past. In the disturbed state of Greece, 

which we should naturally expect the from the death of Alexander b.c. 323 

utmost uniformity to prevail, while as to the establishment of the Greek 

it gave the character to all Greek lite- empire, a.d. 395, took place most pro- 

rature, it became the norm for all bably those changes which separate the 

"correct" speakers. England offers modern from the ancient system, 
the utmost contrast to this state of 

things, and the violent succussions of ^ I papiri diplomatici raccolti ed 

two civil wars, the forcing of a peasant illustrati dall' abate Gactano Marini^ 

into a court dialect, the adoption of a primo custode deila Bibl. Vatic, e pre- 

whole vocabulary from a foreign tongue, fetto degli archivi secreti della Santa 

the parliamentary introduction of pro- Sede. In Roma 1805, fol. 

§ 4, No. 1. 



instances it has been corrected by bis facsimiles, bnt the passages 
ought to be carefully re-edited from the original documents. The 
numbers and pages refer to Marini's book, and the numbers in 
( ) to the lines of the document. The Latin contractions have not 
been extended, and Marini is not always clear as to their meaning. 

No. 75, p. 116. Rome, in the Yatican. 

Attestation to a will a.d. 575, by which certain property was 
left to the Church at Ravenna. The numbers are those of the lines. 
Corrected by facsimile, plate V. 

(24). nerpo^ vJi. KoAeKrapiccs ovet 
Tr]araixriuT(»}v poyaros a Mavvawq . . . 

(25) .... TrfaTaTwpr] (piMcas KwiJ-da 
NapSepTj ijil/ow irp-qcrevTr] er <Tov(TKpiv€VTr) 

(26) .... €i T'QffTafj.evTca prjXcKTou 
TTcp Kov Kouserover ep7j5e aaura T]K\i(na 

(27) . . , .Ka PavevvaTT] rrjsrjs (xovff- 

IS'o. 90, p. 139. In Bologna, Museo dell' Institute. 
Deed of Gift to the Church at Ravenna, vi th or vii th century. 
Corrected by facsimile, plate XII. 

(38) Mapinos xp^cc^KaraXaKris oveiK (38) Marinus Chrysokatalactis huic 

(24) Petrhs vli Collectarius huic testa- 
mentum rogatus a Mannane (25) vd 
testarore filio qd. Nanderit ipso prae- 
sente et subscribente (26) adque ei tes- 
tamento relictum per quod constituit 
heredem santam ecclesiam (27) catho- 
licam Eavennatem testis subscripsi. 

XapTov\e ovaovcpoprv . . . (39) nariunis 
ssTapovfi (re| ovUK^apovfj. TrpiKnrapia) 
mnirp . . . (40) nofxumre tcotiovs aovs- 
rarie fiovueAe er infiooitXe s ... (41) 
fjLWfieinrniovs criyKovii aovirepiovs Aeyt- 
Top (paKTU .... (42) suKTU PaUevuaT€ 
E/cAto'te a looanne uk EiSTrarap . . . (43) 
T^opyi Ma€L(Trpo MiXir . . /x ct nonov 
Upi/xiKipiovs JSovjx . . . (44) Kovfi . . . 
.... /c . , . . Scouarovpe koi fxi Trpeffe 
. . . (45) nov saKTi KpoKLs (piKer . . t 
Kcopa nous ei piXiKTa . . . (46) ros au 
eoSe/j. resTis sovsKpL\l/L er Se KOficrip . . . 
(47) niuovs ooixinuovs K6 sovir^piovs 
svKKpira K^yovn . . . (48) saKra evua- 
yeAAia KopiroppaKir^p /jL^ei irpesevri . . . 
(49) , . . M ovK Trepnoweware saKre 
Vauennare e/cAtste rpa . . 

chartulse usufructuarioe (39) donationis 
sstarum sex unciarum principalium in 
integro. super (40) nominatae totius 
substantias mubilae et immubilac seseque 
(41) moventibus sicut superius legitur 
facta in sstam (42) sanctam Kavennatae 
Ecclesiae a Johanne vc. Expatario qd, 
(43) Georgio Magistro Militum et nunc 
Primicirius Numeri felicum (44) qd. 
Theodosiakus ssto donatore qui mi 
presente (45) signum sanctae Crucis 
fecit et coram nobis ei relicta est (46) 
rogatus ab eodem testis subscripsi et de 
conservandis (47) omnibus (?) omnibus 
quae superius superscripta (?) le- 
guntur ad (48) sancta evangelia cor- 
poraliter mei presentia [praebuit sacra- 
menta et banc donationem] (49) ab hoc 
prenominatae sanctae Eavennatae Ec- 
clesiae traditam [vidi]. 

ITo. 92, p. 142. Rome, in the Yatican. 
Deed of Gift, vi th or vn th century. Corrected by facsimile, 
plate XIII ; line 19 is scarcely legible, and the whole is very obscure. 

(17) (l>n s€(panos iXXovspios Konfianens 

(18) eu KL^nare NeairoAiTavae oik [/cap-] 

(20) TouAe a die Trpeaevri dona .... 

(21) de aoirpa KTKpnrra ofxnia enfiofiiKia 

(22) TTpedia kvi (tovht reppiTopio Ayov 
(23)[Btn]o oujSi ovfii aeov enrpo Kifiirare 

(24) [o-ejou (popL KL^LTttTe lovpis fj.ei a fxe 

(25) (paKTe en ffariKra eKKXeffiaVa^en 

(26) naTe ad ofxnia (TOTrpaiffKpnvTa pe 

(27) Keyi Konarencri er (TovcrKpir^i er reses 

(28) Kvi aovffKpifiepeuT poyafii. 

(17) En Stephanos illustrius conma- 
nens (18) in civitate Neapolitanae huic 

(20) cartulae a die pra?sentis donationis 

(21) de supra inscripta omnia inmobilia 

(22) praedia quae sunt territorio Agu- 
(23)bino ubi ubi sen intro civitate 

(24) seu foris civitate, juris mei a me 

(25) facte en sancta ecclesia Raven 

(26) nate ad omnia suprainscripta re- 
(27)legi consensi et subscripsi et testes 
(28) qui subscriberent rogavi. 



Chap. V. 

No. 110, p. 1C9. Eergamo, in the possession of the Marchesa 

Antonia Solzi Suardi. 
Deed of Gift. Supposed to be of the vi th century. No facsimile. 
(9) . . , . a-ir. ovi ovaocppUKTuapiai (9) . . . sp. huic usufructuariae do- 

dovaT^ioves KaprovKai nationis cartulae ssti hortus in integro 

(11) (6 iriiTei qui est in porgulis cxornatus cum usu 

aOKOiai ivypeffo e0 eypecro vey vov . . 
(12) , . . ptere o TrAarea veK ofjLvi/xovs ad 
fodf/x Tr€pT€i/e ... (13) ... aiKod air. 
\eyiTop (paKTu a a"n. yavOiUcro pev . . . 
(14) . . . fio de(pei'aope cavre eKKAecriai 
pavevvare ... (15) . . . iv crtr. pavivva- 
refi e/c/fAetr k jx . . . (16) . . . e a^i(rpi\ptd 
€0 Kopav V ... IS at peAiKra ear . . . 
(17) ... over a air. yavQiti<ro reCTes 
ffHa-Kpirpi ed avK ... (18) ... Aa^u 
irocriTa (rovirep aaura evayyeAia ukt^io 
... (19) ... prepare e/c/cAeCiat a fieixopaTo 
yav(io(To aovK . . . (20) . . . vpavdo/j. 
rpadeTttfj. vidi. 

No. 114, p. 172. 
Deed of Sale, yi th century. 

(92) ISXianos uh. Apyenrapios ets 
€i<XTp(ayicnris iiiy^vrai i^iycpou (93) 
(poiulcL KoriKoopcfiaKOS pooyaros a 6op- 
^iKionai o<p. fiarpe (9-i) er afi eiocTKoi. 
(})i\iiis dofineKa ocp. er devrepio uh. (X(T 
... (95) ... indiTcapeiios enrais Tvpeaenre- 
fiovs Tcaris aoaKp ... (96) \pi er aa: 
irperio avpi aoAedos Kenrov dcKei eieis 
ey TTp ... (97) euTia rpaderos uidi. 

cortis (11) et putci adque ingresso et 
cgresso ncc non et (12) pariete vel om- 
nibus ad cundem pertincntibus. (13) 
sicut sp. legitur facta a sp. Gaudioso 
reverentis(14)simo defcnsore sanctse 
ecclesia? Ravcnnata) dona(15)tori in sp. 
Eavennatem ecclcs. qui me (16) pre- 
sente subscripsit et coram nobis ei re- 
licta est (17) rogatus a sp. Gaudioso 
testis subscripsi et banc (18) cartu- 
1am positam super sancta evangelia 
actionariis (19) prefate Ecclesiae a 
memorato Gaudioso sub (20) jusjuran- 
dum traditam vidi. 

Borne, in the Yatican. 

No facsimile. 

(92) JuHanus vh. Argentarius his 
instrumentis viginti jugerum (93) fundi 
Concordiacus rogatus a Thulgilone hf. 
matre (94) et ab ejusque filiis Domnica 
bf. et Deuterio vb. sstis (95) vendi- 
toribus ipsis prsesentibus testis subscri- 
(96)psi et ss. pretium auri solidos cen- 
tum decem eis in pr8es(97)entia traditos 

No. 122, p. 187. Eome, in the Vatican. 
Deed of Sale. a.d. 591. No facsimile. 

(78) UuK^KpLKos BA . €is esopfienris 
(re| en inrpiypo ovDK^iapovfx <povn^i 
TeneKeiani (79) cikot covirepicos 
\eyiTop poyaros a aa. VovsiK^iaua h(p. 
tienderpiKai €iova{80)Kae lovyaXh 
Keirane ^d avrovpe ed ecrironraneo} 
(pediiovffffovpe Koe (81) /xe irpeaenre 
aiyna (peLKaepovfi. cd eeis peAi/cro eo* 
resTis crovaKpixpi (82) er crovwpaeffKpnrTO 
TrpeKeiw avpi aoAidos tteieUTi Kiivrovp 
eets €W 7rp6(83)cr£«Tm luianne BK. 
KomrapaToope arno/xipaTOS er rpadiTos 

The Latin A is here uniformly represented by a. But E, though 
generally e, is often ?;, and yery rarely i, indicating not so much 
a wavering pronunciation of e, rj, c, as an uncertain appreciation 
of the sound of the Latin e, confirmed by modern Italian usage. I 
is regularly t, but not unfrequently et ; in tuyevrac viginti (No. 
114, line 92), if the transcription is to be trusted, l, e, au all occur 
for % and e is also found occasionally, compare ueienrc (No. 122, 

(78) Pacificus vh. his instrumentis 
sex in integro unciarum fundi Gene- 
ciani (79) sicut superius legitur rogatus 
a ssta. Rusticiana hf. venditrice ejus- 
(80)que jugale Tzitane vd. autore et 
spontaneo fidejussure qui (81) me 
prsesente signa fecerunt et eis relictum 
est testis subscripsi (82) et suprascrip- 
tum precium auri solidos viginti qua- 
tuor eis in pr8e(83)sentia lohanne vc. 
comparatore adnumeratos et traditos 


line 82) ; this again must be attributed to mishearing of the Latin. 
is 0, CO, and rarely ov, for similar reasons. U is regularly ov, 
occasionally o, v in the words, koc, kvc, for qui, and rarely co. I 
have already recorded my opinion that the original sound of Greek 
ot was (ui), and Latin oe (ue), see Trans. Phil. Soc. 1867, supp. p. 65. 
Probably adKoiat = atque (No. 110, line 11) is Marini's misprint for 
adfcovat, AE is generally e, occasionally at,, AU is represented 
by av in avpc = auri, No. 122, line 82, but it is stiU possible that 
the Greek said (abh*ri), as I heard a guide at Pompeii call centauro 
(tshentabh-ro), and compare PavevvaTr}= Ravennatem. The Greek et, 
OL are written occasionally for el, o'i; compare ei9, eei<; = eis, oveiK oik, 
=liuic. Among the consonants /5 is used for Latin v =(bh) ?, and h, 
but Latin h is also represented by «< a special form of i; ; 7 is used 
for g which, however, occasionally falls into t ; S is rather avoided, 
or receives a special form d for Latin d; ^ only occurs in one of the 
attempts yav^ioao to spell Gaudioso, and in aKT^co, dovaT^iov€<; 
for actio donationes, which seem to indicate its present use in t^, vt^ 
= (tsh, dzh), but observe the pure t in irpeacvrca = prcesentia; is 
only used as a mispronunciation oi t; k universally represents c, 
indicating that the Latin letter had preserved its sound down to 
this period in Italy, as indeed the ags. use of c is sufficient to prove ; 
X = / • jM = m, but the m is often quite dropped when final, indi- 
cating the transition to the modern Italian -0, -a, from -um, -am, the 
accusative forms ; v = n, but n and m are much confused ; ^= x^ 
TT = p, p = r, (T = s, T = t, <p = f, X ^^^^ ^^^ occur, ^fr = ps as in 
Tjyirov = ipso, aovaKpi'^^n = subscripsi, but etTrat^ = ipsis, is also 
found. The use of aavra = sancta, seems to indicate a transition 
to the modern Italian santa, although aaKra, aanKTa also occur, 
and the combinations 77, 7/c are not found. 

The extremely recent date of the present pronunciation of Greek 
in England is not generally appreciated. In 1554 the present 
modern Greek pronunciation was regularly taught.^ Sir Thomas 

^ See: Institutiones Lingvse Grsecae ; grec, je I'ay entendu. Et comment.? 

N. Clenardo Authore cum Scholijs as tu demeure en Grece ?" The Greek 

P. Antesignani Rapistagnesis, Liigduni, is thus restored in the edition of the 

1554, in which the only pronunciation ffiuvres de Rabelais par Esmangart et 

taught is that now usual at Athens. 'Eloi Johanneau (Paris, 1823, 9 vols. 

Compare also the passage in Rabelais 8vo.) vol. 3, p. 296. AecrnoTa roiuvv 

— La vie de Garagantua et de Panta- iravdyade, Sia t\ av fioi ouk aproSoreTs ; 

gruel. Book ii, chap. ix. (first edn. bpS,s yap Xi}xcf avaXiaKoinepou ifxe 

1535), " Dont dist le compaignon : ddXiou, koI iu reS fi^ra^u /ue ovk eAeels 

" Despota tinyn panagathe diati sy mi ovSafxws- ^Tjreis Se irap' i/jLov a ov xp^- 

ouk artodotis F horas gar limo analis- Kai 'dfioos (piXo\6yoi Trdures ofxoXoyovai 

comenon erne athlion, ke en to metaxy rore Xoyovs re koI prjfxara ir^pLrra 

me ouk eleis oudamos, zetis de par inrdpx^iy ottSts irpajfia avrh iraai 

emou ha ou chre. Ke homos philologi Sr]X6u iaTiv. "EvQa yap avayKoioi 

pantes homologousi tote logons te ke fxSvou Xoyoi ^1(t\v, 'ba izpdyixara, Sju 

remata peritta hyparchin, opote pragma irepi a/j.(pia^r], ^ut? irpoatpopus eVt- 

afto pasi delon esti. Entha gar anan- (paivTiTai. Observe the retention of e 

kei monon logi isin, hina pragmata for 77 ; dialectically (riSepou depiou, etc., 

(hon peri amphisbetoumen), me pros- are still found for (xidripov drjpiou, etc., 

phoros epiphente." Quoy.? dist Car- in Modern Greek, 
palim lacquays de Pantagruel, c'est 34 



Chap. V. 

Smith's theories were quite heretical in 1568, see supra, p, 35, 
note 1, and he called a, e, t], i, o, (o, v, at,, ei, av, ev, ov, vl (aa a, 
e, ee, ei i, o, oo, yy, ai, ei, au, eu, ou, wei), entirely ignoring the 
long sound of (ii) both in Latin and Greek. In the xvii th century 
a, I, V, ei, av, ev, ov, became (seae se, oi i, iu, oi, aa, iu, au), in 
the XVIII th a, rj, became (ee, ii), and thus in one letter, 77, the 
former pronunciation was restored. The extraordinary mispronun- 
ciation of Latin and Greek now prevalent in England, results from 
the application of our own changeable pronunciation to the fixed 
pronunciation of dead languages, and from the historical ignorance 
which assumes that a language may have only one pronunciation 
through the generations for which it lasts. We may never be able 
to recover the pronunciation, or appreciate the quantitative rhythm 
of the Athenian tragedians or of the Homeric rhapsodists, but we 
can read as Plutarch and as Lucian, and we should be satisfied with 
that privilege, remembering that if we pronounced these later 
authors otherwise than as the modern Greeks, we should certainly 
pronounce wrongly. It would indeed be just as absurd to read 
Lucian with the pronunciation of Aristophanes, as to read Tenny- 
son with the pronunciation of Chaucer.^ 

^ The following is Kopari 's eloquent 
apology for the modern Greek pronun- 
ciation in the preface to his edition of 
Isocrates, Paris, 1807. No one who 
is acquainted with ancient Greek will 
have any difficulty in reading it, and 
the English pronunciation of Greek is 
so mixed up with the history of our 
own pronunciation, that it is not out 
of place to give it here at length : — 
'Xu^oi/rai TvoXkorarai ^iriypacpal ira- 
Xaial, Tuu diroicov 7] KaK^ ypa<pri ano- 
Seix^^h OTi Twi/ ayifxcpivoov 'EWrjuoov 
rrjs 'EK\7]PiKris yXccaar]s r/ irpocpopa ehat 
7] aurr? Kai 7] vpocpopa, ^tls i)Tov els 
XpVdiu Kara robs KaiaapiKOVs, Kol taws 
avairepa kut' aurovs tovs UToXefiaiKOvs 
Xpovovs, ^yovv kut' iKeij/rjv o\t}v ttju 
irepioSov tov xP^^^^i -'^ '^^^ oiroiav 
€^r](rav Kade^ris 6 UoXv^ios, 6 'hXiKapv • 
acraevs Aiovvcnos, 6 ^iKeXiuTrjs Aio- 
dcapos, 6 %Tpdli<jov, Kal 'av eXOcvfieu 
KarcoTepw /xe'xpt T'^s SevTepas anh 
XpLarou eKarouraeTTipiSos, Aicau 6 
Xpva-ocTTOiJLOs, o UXovrapxos, 6 'kppia- 
vhs, b Ylavcravias, 6 AovKiavhs, 6 TaAij- 
vhs, 2e|T0S 6 'EfiireipiKhs, Kol &Woi 
iroWol a^iSXoyoi avyypa(pi7s. '"Eaj/ 
^vai fidplBapos rj cn^/xepiVT] 7]fi(vv irpo- 
(popa, eiv' iKelvoi ox! vfieh ot atrioi 
ttJs fiap^apuaecos," ifxiropovixey va 
aTTOKpivwixeu irphs tovs Karrjyopovs, Ka\ 
VOL TOVS TTapaKaXeffcc/xeu va vTro<p€p(a<nv 
fi€ ixaKpoOviJLiav vo Trpocpepcofxev Kal 

rjixus^ cos iirp6<pepav e/ceTi/ot. '^TTjpiCeTat 
fxaXiaTa 7} KaT7)yopia els tov 'lajTa/cto"-' 
^lbv, ijyovp T^v i^avdyK7]s ffvjxfiaivov- 
aav TOV avTOv tJX^^ "^^^ 'laJra avxf^v 
iiravaATixl/iv, dTrSrav Kal at SicpOoyyoi 
EI Kal OI TTpo(pep(avTai uis aiiTS. 'Afx(pi- 
fioAla Sev eJvai on t] crvx^'r] toov avToov 
aToix^iuiv €Travd\7]\pLS elvai (pvaiKo, 
arjSrjs- aAA' o^i Sia tovto TrpeVet tis 
iravTOTe va t^v dTro(pevyri /ie Trepiepyiav 
SeiaiSai/xova, OTav ixdKicTTa 8ev ^vai 
av/xcpcova TO, iiravaAafifiavoixeva CTOi- 
X^^'^ j Hapa^eiy jxaTOs X^P''^ ^^^ Thvi 
arixov tovtov tov 'OiJ.'f]pov ('lAtaS. E. 

Oioi Tpdo'ioi 'Itcttoi, iirKTTdfievoi 

evpiffKeTai i^dKis 7] Slcpdoyyos 01. M'l 
'6\ov TOVTO Sev )8A67rco dia iroiav alTiau 
irpofpepofjievos KaTo, tt/j/ irpocpopav twv 

u Tpdo'u '{tttti, eTTifTTa/ievt ireSiio 
ijOeXev eJadai els T7]v aK07]V ar]5eaTepos 
Trapa irpocpepo/xevos, cos Thv Trpocpepovffi 
TToAAoi airh rovs aWoyeveTs EvpaTralovs. 

b'io'C Tpca'io'i 'linrot iTTiffTdfievoi TreSio'io 
Se'lTOs 6 'EfJiireipiKhs ovoixd^ei KaOapa 
Tas Si(j)66yyovs TavTas (XToix^^a, ^yovv 
Tas ffToxdC^Tai ws airXa ypd/xfiaTa els 
T^j/ Trpo(popdv. [In a footnote the 
author says that Sextus lived a.d. 190, 
and cites a long passage from his Uphs 
Tpa/jL/iiaTiK. Keep, e, § 117, treA. 241, 
beginning : 'ETrel oi/v 6 tov AI Kal EI 

J 4, No. 1. 



After thus establishing the value of these transcriptions of the 
Septuagint into Anglosaxon characters for indicating the precise 
signification of the Anglosaxon vowels in the x th century, it may 
seem superfluous to cite J^orman traditions in the xn th and xm th, 
were there not always a certain amount of satisfaction in cumula- 
tive evidence. In Wace's Eoman de E,ou, which unfortunately 
exists only in later transcripts, and whose author probably always 
pronounced the despised Saxon most vilely, and certainly spelled it 
abominably, we find the following indications. Describing the 
conduct of the Saxons the night before the battle of Hastings, he 
says : 

Mult les veissiez demener E laticome e drincheheil. 

Treper e saillir e chanter Driyic Hindrewnrt e Drintome 

Biifler e crier welseil Drinc Keif ^ drinc Tome. v. 12471-6 

which may perhaps be rendered: ''You might see them much 
sporting, gamboling, leaping, singing, joking,^ and crying Tr(2s 
Jicel, and Lmt hit cuman, and Drinc heel, Drinc Hindweard, and 
Drinc to me, Drinc healf and Drink to me.'''' In this Wees heel and 
Driiw heel are well known, and we must not be surprised at finding 
IsTorman ei for ags. ^, a strange sound, when Orrmin shews he'^'^tenn 
for ags. heatan (supra p. 489). Drink to me, remains in our language. 

<pd6'Yyos aivKovs icrri Koi iiji.oj/oeid7]s, 
fcrrai koI ravra (rTotX€?a, and proceed- 
ing very distinctly to shew that by this 
expression he excluded the conception 
of diphthongs.] Kal au tovto Sev 
aiTodeixvy on els rovs xp(^i/oiis tov 
2e|Tou 7} ■jrpo(popa deu ^to c()6apfji.ei'r]y 
'iKavov eiuai va Sei^rj, on els tovs xpof^ovs 
TOV 5eu virooirreveTO KUPels^ oti ol 
o\iyas eKaT0VTaeT7]pi5as Trpoyevearepoi 
e?Xaj' irpo^opav Sid(popov. Mtjt' i^evpco, 
/iTjre va jua0co ^e /xeXei, irws iirpocpepei/ 
6 l(T0KpdT7}s, d nAaTWJ/, A7]fj.0(r6euris, 
Kol oaroL aWoi ^Kfj.aaav els avTTji/ ttjs 
yXct}(ra7]s ttjv aKixTjv Koi, orav virepa- 
(TTTL^to TTju (n)jx€piv7]v TTpocpopau, Seu 
Sti'frxyp'Coftct' oTi Trpocpepoixev airapaK- 
KuKTccs ws eKeipoi, eirei5r] iriOavov elvat 
va ecrvve^f) Ka\ els tt]u 'EWriuiurji', o,n 
(rv/xfiaivei els o\a rwu duOpwircou ra 
epya Kal TrotTj/xara. Tovto /j.6uov aSicr- 
raKTcas Tncrrevu} on av r] irpocpopa tyis 
yXdoaarjs riXXoLcJciOr], va t)]V avoKara- 
(TTTiaj) els TTjv apxaiav avrris (pvaiu Sev 
elvai KaXhs Trapa jx6vol ol 6iro7oi rrjv 
e\d\ovv Kal Tr?j/ eypacpov cos luLTjTpiK-rjv 
avTwv yXooaaav. "Ecos va avalSicoawaiv 
eKelvoL, Kal els crvyxooprjfxevov 
eJvai va Tvpocpepcofjiev, cos ttji/ iirp^cpepev 
6 fidp^apos Se'lTOS, o aypd/x/jLaros YlKov- 
Tapxos, 6 afxaOeaTaTOS Takrivos, Kal ol 
aWoyeve7s 'EWr^viaral cpiXocrocpcJcTepov 
TjQeXav irpd^ei, av eirefxivov Kal ttjv irpo- 
(popdv TOV 'Epda/xov ottov eirefMirov ttoA- 

}[as aWas rrpoX'fjxpeLS, Tcopa ^idKicrTa 
els TT]v avayevvif)(nv ttjs 'EAActSov, 
oTrSTav fie ttjv o/xocpooviav tTjs Trporpopas^ 
Kal Tr]v aSiaKOTTOv irapdOeaiv Trjs 
iraXaias fxe t))v veav yXwaaav tSjv 
'E\Ar]vcov, Kal avTol dirh Tas aKSjUf) 
SeiAas rjfxciv TrapaTriprjo'eis, Kal ri/xels 
airh Tas aocpas avTcuv arjfxeLcvaeis ride\a-- 
/xev fxeydXcos w(pe\r)drj els Tr]v KaTa- 
v6r]a'iv Tuv apxaiav iroiriTuv Kal avy- 

1 I adopt the reading of the Duchesne 
MS. cited by Pluquet, since the read- 
ing in his text "£ubite crient e weissel" 
is unintelligible. Bnfler'\'& from ^^buffe, 
buffet, bvffle : coup de poing, soufflet, 
tape ; huffa, en Ital. buffettone ; en 
Basq. bufeta ; en Languedocien bnfa," 
(Roquefort) ; whence English buffet y 
compare Italian buffo, whence our buf- 
foon. Compare also the Norfolk buffle, 
to handle clumsily, to speak thickly 
and inarticulately (Nail), to abuse, to 
rate soundly (as I am informed by Mr. 
Waring) ; also German Biiffel, buffalo, 
buff, lout (compare Ochs for a fool) and 
biiffeln to drudge (Hilpert). Whether 
buffer is a Norman word adopted into 
English, or an English word Norman- 
ized — compare the modern French 
boxer, to box — it is impossible to deter- 
mine in the absence of parallel passages. 
It seems here to imply rough joking. 


Perhaps L<^t hit cumaUy is a good wish, may you have what you 
want, and the diinking hindweard and healf, may refer to some 
customs such as still prevail among those who, making an art of 
toping, such as standing back to back and giving each to drink from 
the other's cup, or both drinking from the same bowl, etc. Th* 
passage is, however, not of much service phonetically, and the 
Angiosaxon words are doubtful. The following are better : 

Olicrosse so vent crioent, E Godemite altretant 

E Godemite reclamoent : Com en frencciz Dex tot poissant. 

0/i{?>-05At' est en engleiz v. 13119-24. 

Ke sainte Croix est en franceiz, 

Hence Olicros&e = Hdlig Cross, which looks like an error for Rod, 
and Godemite is God Almihtig. The former would incline to a 
very broad pronunciation of a as (aa), and perhaps arose from the 
subsequent southern Jiohj. The latter might imply that long i was 
(ii), and certainly that they did not pronounce almighty as at pre- 
sent ; but as the vowel was certainly short in miht, we do not gain 
much, except to learn that this form coexisted with Orrmin's 
AUmahhti'^. The form Godelamit occurs in the singular poem 
called La Pais aux Englois, attributed to a.d. 1263, which ridicules 
English Prench in an orthography difficult to comprehend.^ 

Normanz escrient : Dex aie ; Con est I'ensegne que jou di 

La gent englesche : JJt s'escrie. v. 13193 Quant Engles saient hors a cri. 

The two last lines are an addition to the text of Pluquet, taken 
from MS. 6987, Bib. Roy. de Paris (E. Taylor's translation, p. 191), 
and imply that ut = ags. ut, and therefore fixes the traditional pro- 
nunciation as (uut), which is of some value. The Man of v. 109, 
and Zoonee of v. 10659 (supra, p. 461, note col. 1) arc useless. 

Marie de France belonged to quite the beginning of the xinth 
century, and we have the advantage of an indubitably early manu- 
script of much of her poetry.^ In her lai de Laustic (Poquefort 1, 
315, Harl. MS. 978, fo. 142), which Eoquefort explains as in- 
tended for a Breton word, meaning a nightingale, she says : 

Lauftic ad nun ceo meft auif Ceo eft reifun en fr«nceif 

Sil apelent en lur paif E nihtegale en dreit engleif. v. 3. 

1 See Journal de 1' Institute His- consistent way in which dialectic or 

torique, Premiere Anne, 1834, p. 363, foreign pronunciation is still repre- 

for which reference I am indebted to sented orthographically, e.g. Burus's 

the kindness of M. Francisque Michel. Scotch. No doubt can be felt as to 

In this poem roi is uniformly spelled the presumed rhyming word fab-e (p. 

rai^ and foire rhymes to Ingletiere, 449), after seeing Orrmin's ortho- 

gmre, conquerre^ which seems to mili- graphy fa^^err, p. 489. 
tate against the view I have taken on ^ xhe Harl. 978 described supra, p. 

p. 453, and at least shews that (feere) 419. The Fables of Esop there named 

was a presumed Anglo-Norman pro- are by Marie de France, and many of 

nunciation at the time, but whether it her lays occur in the latter part of the 

was the only or general value, or same MS. See : Poesies de Marie de 

whether this may not be due to the France, poete Anglo-normand du xiiie 

author's pronunciation, or to the Poite- siecle, par B. de Moquefort^ Paris, 1819, 

vin dialect to which the editor attri- 2 vols. 8vo. I am indebted to Mr. 

butes the piece, it is difficult for any Payne for having drawn my attention 

one to determine, who knows the in- to the transcription of English in her 


In the lai de Chevrefoil (Eoq., 1, 388, Harl. MS. 978, fo. 148J), 
we find : 
En fuhtwales .v. il fu nez v, 16. Gotelef lapelent en engleif 

En cornwaille uait tut dreit. v. 27. Cheurefoil le numewt en f?-«nceis. v. 115 

In the lai de Milun (Roq. 1, 328) we find Suhtwales v. 9, 
Irlande 15, IJ^orweie 16, Guhtlande 16, Suhthamptnne 318, I^or- 
thumbre 453. In the lai d? Ytvenec (Roq. 1, 274), we have Incolne 
= Lincoln v. 26, and Yllande = Ireland, v. 27. In the Eables 
(Roq. 2, 141, Harl. MS. 978, fo. 53h), we have: 

Si ad ure ke li uileinf Lung cum li witecocs aueit. 

Euft tel bek mut U plereit v. 18-20 

where Roquefort cites the variants : huitecox, widecos, witecoc, 
which all seem to mean wMtecock, an unknown bird, but as Norman 
m was probably not so truly (ui) as (ui), or according to Mr. Payne 
(uu), p. 424, n. 3, and certainly often replaced (uu), p. 458, 1. 27, 
these may mean (uit'ekok, uut'ekok), that is (wuud'ekok), ags. 
wuducocc (Ettm. 86), English woodcock, with an omitted (w) 
before (uu), p. 420, note, col. 2. These words give (aa a, ee e, 
ii i, 00 0, uu) as Marie de France's appreciation of the sounds of 
the Anglosaxon, or xii th century English a, e, i, o, u. 

In order to see at a glance the different opinions that prevail 
respecting the values of the Anglo-saxon letters, a table has been 
annexed on p. 534, giving also the views of Rask, Grimm, and 
Rapp.^ JSTeither Rask nor Rapp give any illustrations, though Rapp 
writes a few isolated words. ^ But as we have ventured to give a 
theoretical representation of the values of the letters, symbolizing 
of course difi'erent pronunciations according as they are used in 
different combinations to express the very distinct dialects which 
prevailed at the time, it is necessary to shew the effect of this 
theory, by attempting the phonetic representation of a short passage. 
The parable of the Prodigal Son,^ has been selected for this purpose, 
and will be hereafter presented in Icelandic (No. 2), Gothic (No. 3), 

and Wace's poems. It is true that her ii, 140-149, iv, 245, Yergleichende 

transliterations of English rather repre- Grammatik, vol. 3 (1859), pp. 125-129. 

sent the pronunciation of the xiiith ^ This being contrary to his usual 

century, than of Anglosaxon, and should, custom he explains by saying: "Da 

properly speaking, have been adduced dieser Dialekt noch zu gar keinem 

on p. 462, but as I was not aware of festen Eesultate iiber die Kritik der 

them till after that sheet was printed Buchstaben gelangt ist, sind wir weit 

off, I am glad to have this oppor- entfernt, mit dahin einschlagenden 

tunity of inserting them. Sprachproben uns zu befassen." 

^ £. Rask, Grammar of the Anglo- ^ £)a, halgan Godspel on Englisc. 
Saxon Tongue, translated from the The Anglo-Saxon version of the holy 
Danish by B. Thorpe, Copenhagen, Gospels, edited from the original manu- 
1830, pp. 6-15. /. Grimm, D. G. I^, scripts by Benjamin Thorpe, F.S.A., 
325-378, for vowels, and I^, 243-269 London, 1842, 8vo. pp. 240. "The 
or consonants, but the indications are basis of the present text is the Cod. 
often so indistinct, that much doubt is Bibl. Pub. Cant. Ii. 2, 11, collated 
to be attached to the following inter- with Cod. C. C. C. C. S. 4. 140. In 
pretations. Grimm proceeds from an doubtful cases Cod. Bodl. 441. and Cod. 
etiological, rather than a phonetic Cott. Otho, C. 1, have also been con- 
conception. K. M. Rapp, Phys. d. Spr. suited." — Preface. 



Chai'. V. 

the Wycliffite version (Chap. VIL, § 3), for the sake of comparison. 
The translation at the foot of the page is intended to point out the 
^ammatical construction, and the etymological relations of each 
word to the English, and would be therefore scarcely intelligible if 
the passage were not so well known. 

























a a 













le ie 




























k k 













q qg 


























e e 

e e 







saa ja 

ea e&. 

ea, ea 

ea ed 


sk sk 


sk s^ 



eo eo 

eo eo 

eo eo 






f V 



f V 







g ff J 

H' kb 



g J 

g ff 





























yy ii 

y *■ 

Anglosaxon, Lucas 15, 11-32. 

11 So^lice sum man haefde 
twegen suna. 

12 Da cwae^ se gm^TS,\Thorpe, 
yldi'a] to his feeder, Fieder, syle 
me minne dael minre sehte \q 
me to gebyre^. Da daelde he 
hym hys sehte. 

13 Da, sefter feawa dagiim, 
ealle his fing gegaderode se 
gingra sunu, and ferde wrseclice 
on feorlen rice, and forspilde 
far his sehta, lybbende on his 

Conjectured Pronunciation. 

11 Soodh'lii^e swm m«n 
H86V'de twee^h'en swn*a. 

12 T1m<? ke^aedh se ^h«q*ra 
to h/s faed'er, Faed*er, sybe me 
miin*e daeael miin're aekht'e thee 
me too'gebyr*edh. T\iaa daeaeld'e 
He H/m H«s 8e8eZ;ht*e. 

13 Th«rt, aeft'er fea'wa dagh*- 
um, eabe h«s th/q gegad'erode 
se ^h^q*ra swn'u, <?nd fer'de 
re^'se/j'lii/ve on feorden rii^*e, 
and forsp^l•de tha^r h«s aeaekht'a, 
lyb'ende on h/s gsesebsan. 

Verbatim Translation^ Luke 15, 11-32. 

11 Soothly some man had twain sons. 

12 Then quoth the younger to his 
father, Father, sell (give) me mine 
deal (part) of -mine owning that me to 
belongeth. Then dealed he him his 

13 Then, after few days, all bis 
things gathered the younger son, and 
fared banish ed-like (abroad) on far 
kingdom, and for-spilled (lost) there 
bis ownings, living on his luxury. 

§ 4, No. 1. 



14 Da he hig hsefcle ealle 
amyrrede, j^a wear^ mycel hun- 
ger on jjam rice ; and he wear^ 

15 Da ferde he and folgode 
anum burh-sittendum men j^aes 
rices : fa sende he hine to his 
tune, f set he heolde hys swyn. 

16 Da gewilnode he his 
wambe gefyllan of j^am bean- 
coddum ]7e ^a swyn seton : and 
him man ne sealde. 

17 Da befohte he hine, and 
cw8e^, Eala hu fela yr^linga 
on mines fgeder huse hlaf ge- 
nohne habba^, and ic her on 
hungre forweor^e ! 

18 Ic arise, and ic fare to 
minum faeder, and ic secge him, 

19 Eala faeder, ic syngode on 
heofenas, and beforan j^e, nu ic 
neom wyr^e J^aet ic beo Jin 
sunu nemned : do me swa aenne 
of J7inum yr^lingum. 

20 And he aras fa, and com 
to his faeder. And fa gyt, fa 
he wfes fcor his faeder, he hyne 
geseah, and wear^ mid mild- 
heortnesse astyred, and agen 
hine arn, and hine beclypte, and 
cyste hine. 

14 Thaa He mffh naevde 
eal'e amyr'ede thaa weardh 
m«^"el TLuq^-ev on th«m riiy?;*e ; and 
He weardh waed'la. 

15 Thaa fer'de ne and. fol*- 
ghode aawnm huikwh-sit'endum. 
men thaes rii/^'es : thaa send'e 
He Hm*e to H2s tuu'ne, thaet ne 
Heold'e H^s swiin. 

16 Thaa gew^I■node ne h«s 
wam'he gefyl'an of tham. bean*- 
kod'um thee tha swiin aeaet'on : 
and H2m m^^n ne seal'de. 

17 Thfl^a bethokht'e He Hm*e, 
and kz^;aeth, Ea"la, huu feba 
yrdh'l^qa on mii'nes faed'er 
Huu'se lh«^f genookh'ne H«;b'- 
ath, and i'k neer on Hwq're 
forweor'dhe ! 

18 rk arii'se, and ik far'e 
to mii'm^m faed'er, and ik seye 

19 Ea-la faed'er, «k syn-gode 
on Heo'venas, ^nd befor'an 
thee, nuu «k neom wyrdh'e thaet 
«k beo thiin swn'u nem'ned : 
doo me swaa aen'e of thii'num 

20 ^nd He araas' thaa, and. 
koom to ms faed'er. And thaa 
^h2t thaa ne waes feor his 
faed'er, He hm'e geseakh* and 
weardh m^d m^ld-heort'nese as- 
t^r'ed, and agen* nm'e am, 
and Hm'e beklyp'te, and kys'te 

Verbatim Translation. 

14 Then (when) he them had all 
dissipated, then worth (became) muckle 
hunger on that kingdom ; and he worth 
(became) destitute. 

15 Then fared he and followed one 
borough-sitting man of-that kingdom : 
then sent he him to his town (inclo- 
sure), that he might hold his swine. 

16 Then desired he his womb (belly) 
to-fill of (with) the bean-cods that the 
swine ate; and to -him man not sold 

17 Then bethought he him, and 
quoth, Oh ! how many earthlings (farm- 
ers) on mine father's house, loaf (bread) 

enough have, and I here on hunger 
forth-worth (perish). 

18 I arise and I fare to mine father, 
and I say to him, 

19 Ob ! father, I sinned on heavens, 
and before thee, now I not-am worthy 
that I be thine son named : do to -me 
as to-one of thine earthlings (farmers). 

20 And he arose then, and came to 
his father. And then yet, then (while) 
he was far-from his father, he him saw, 
and worth (became) with mildhearti- 
ness a-stirred, and again him ran, and 
him be-clipped (embraced), and kissed 



Chap. V. 

21 Da cwae^ liis sunu, Facder, 
ic syngodc on lieofen, and Le- 
foran ^e, nu ic ne eom wyr^e 
J^aet ic Jjin sunu beo genenmed. 

22 Da cwae^ se faeder to his 
Jeowum, Bringa^ ra^e fone 
selcstan gcgyrelan, and scryda^ 
hine ; and sylla^ Wm hiing on 
his hand, and gescy to his fotum ; 

23 And bringa^ an faett 
styric, and ofslea^ ; and . uton 
etan, and gewistfullian : 

24 forj'am ])es min sunu waes 
dead, and he geedcucode ; he 
forwear^, and he ys gemet. 
Da ongunnon hig gewistla^can. 

25 So^lice his yldra sunu 
wses on aecere ; and he com : 
and ]7a he J7am huse genealaehte, 
he gehyrde ]7one sweg and Jset 

26 Da clypode he aenne J^eow, 
and acsode hme hwaet J^aet wsere. 

27 Da cwae^ he, pin broker 
com, and J^in feeder ofsloh an 
fsett cealf ; forj^am he hine 
halne onfeng. 

28 Da gebealh he hine, and 
nolde ingan: ]7a eode his feeder 
ut, and ongan hine biddan. 

21 Thaa kz^^aeth ms swn'u, 
FaedxT ^k syn'gode on neo'ven, 
and befor'an thee, nuu ik ne 
eom wyrdh'c dhaet tkthiin swn'u 
beo genem'ned. 

22 Thaa ktvxth se faed'er to 
n/s theo'wwm, Br/q'adh raadh'c 
thon-e see'lestan gegyr'ehiii, 
and skryyd-adh Hm'c, «nd syl*- 
adh H«m rh/q on uis H«nd, and 
gesAryy to nis foo'twm : 

23 and br/q'adh aan fa^t 
styyriX-, and of'sleadh* ; «nd 
uu'ton et'an, and gew«st*fwl'ian: 

24 fortham* thes miin swn'u 
waes dead, and he ge,edkuu*- 
kode ; He forweardh", and ue is 
gemeet". Thaa on'g«<n*on ui^h 

25 Soodh"liiX*e He's yld"ra 
swn'u waes on aek'ere ; «nd 
He koom ; «nd thaa He tham 
Huu'se genea'laeoekhte. He ge- 
Hyrd'e thon'e swee^/h and thaet 

26 Thaa kly|3'ode ne aen-e 
theou, and aks'ode Hm*e whaet 
thaet waeae're. 

27 Thaa kwaedh ne, Thiin 
broo'dher koom, and thiin 
faed'er of'slookh* aan faet ke«lf : 
fortham* He nm-e na^l'ne on'feq'. 

28 Thaa gebeaU-h* ne nm-e 
and nold'e in'gaan' : thaa eo'de 
h/s faed'er uut, and on'gan* 
Hm'e b«d*an. 

Verbatim Translation. 

21 Then quoth his son, Father, I 
sinned on heaven, and before thee, now 
I not am worthy that I thine son he 

22 Then quoth the father to his 
thanes (servants). Bring rathe (quickly) 
the best garment, and shroud i^clothe) 
him, and sell (give) him a-ring on his 
hand, and shoes to his feet, 

23 and bring one fat steer, and 
slaughter ; and let us eat and feast, 

24 for-that (because) this mine son 
was dead, and he again-quickened ; he 
forth-worth (perished), and he is met. 
Then began they to-feast. 

25 Soothly his elder son was on 
acre ; and he came, and then (while) 
he to-the house neared, he heard the 
music and the company, 

26 Then cleped (called) he one 
thane (servant) and asked him what 
that were. 

27 Then quoth he, Thine brother 
came, and thine father slaughtered one 
fat calf ; for-that he him whole fanged 
(received) . 

28 Then was-wrathful-at he him 
and not-would go-in: then went his 
father out, and began him to-bid. 

§ 4, No. 2. 



29 Da cwse^ he, his fseder 
andswariende, Efne, swa fela 
geara ic j^e j^eowode, and ic 
nsefre ]7in bebod ne forgymde, 
and ne sealdest ]>u me naefre an 
ticcen, j^set ic mid minum freon- 
dum gewistfuUode : 

30 ac sy^^an fes ])m sunu 
com, ]7e hys spede mid myltry- 
strum amyrde, ]>u ofsloge him 
faett cealf. 

31 Da cwse^ he, Sunu, ]7U 
eart symle mid me, and ealle 
mine ^ing synd l^ine : ]>e geby- 
rede gewistfuUian and geblis- 
sian : forj^am j^es J^in broj^er wses 
dead, and he geedcucode ; he 
forwear^, and he ys gemet. 


29 Then quoth he, his father an- 
swering, Lo ! so many years I thee 
thaned (served), and I never thine 
bidding not neglected, and not soldcst 
(gavest) thou me never one kid, that 
I with my friends feasted : 

30 Eke (but) sithens (since) this 
thine son came, that his speed (pro- 

29 Thaa 'kivseth. He, h«'s 
faBd'er andsw<^r"iende, Eevne 
swa fel'a ^^hea-ra **k the theo'- 
wode ; «nd «k ngevre thiin 
bebod' ne for^hyyni'de, and ne 
seal'dest thuii mee naevre aan 
t«k*en, thaet ^k m?'d miin'wm 
freon'dwm gew/st'fwl'ode : 

30 ok siidh'an thes thiin 
swn'u koom, thee His spee'de 
mid mil'tristrwm amyrd'e thuu 
of'sloo/jrh'e H/m fget kedlf. 

31 Tliaa kt^-oedh He, Swn'u, 
thuu eart simde nitd. mee, and 
eal*e miine thz'q s/nd thii'ne : 
thee gebyi-ede gew/st'fwl'ian 
ffnd gebl«s*ian forth«m* thes thiin 
broo'dher waes dead and ge,ed- 
kuu'kode ; He forweardh*, and 
He is gemeet*. 



perty) with mistresses lost, 
slaughtercst for-hira fat calf. 

31 Then quoth he, Son, thou art 
ever with me, and all mine things are 
thine ; to-thee belonged to-feast and 
to-bliss ; for-that this thine brother was 
dead, and he again-quickened ; he 
forth- worth (perished), and he is met. 

2. Icelandic and Old Norse. 

In the IX th century, Iceland was discovered and colonised by 
the Scandinavians. The writing at first used was runic, but 
Eoman Christianity and Roman letters, which seem to have always 
gone hand in hand, were introduced in the xith century, and MSS. 
of the XII th and xmth centuries still exist. The sea usually 
unites; but large tracts of dangerous wintry sea, and a climate 
which for months in the year closes the harbours, separate. The 
Icelandic colonizers were so separated from their native country 
that their tongue was practically unaffected by the causes which 
divided it on the continent into two, mutually unintelligible, literary 
languages, the Danish and Swedish, and the numerous unwritten 
Norwegian dialects.^ In Iceland, therefore, we have the strange 

1 " On the older Eunic stones alto- 
gether the same tongue is found in all 
three kingdoms, and in the oldest laws 
of each people very nearly the same. 
This tongue occurs first under the 
denomination Donsk tunga (Doensk 
tuuq'ga) because Denmark was in the 
oldest times the mightiest kingdom. . . . 

But the Old Norse began also first to 
decay in Denmark, and therefore took 
the name Norrsena (Nor-raarna), be- 
cause it was probably spoken best and 
most purely in Norway .... Before 
the Union of Calmar [between Den- 
mark, Sweden, and Norway, 1397], it 
was materially changed both in Sweden 


spectacle of a living medieval tongue, with all its terminations, in- 
Hections, and vowel changes, whether of mutation ( Umlaut) or 
progression [Lautverschiehung), practically unchanged, and in daily- 
use. The language of the oldest MSS. scarcely differs from that 
of the most modem printed hooks as much as that of Chaucer 
from that of Shakspere. Practically tlie study of Icelandic is 
the study of the language spoken hy those fierce invaders of our 
Eastern coasts, whose tongue has so powerfully and permanently 
affected all our Eastern and Northern dialects. It is, therefore, of 
extreme interest to all students of dialectic or early English.^ But 
its orthograpliic laws are so different from those with wldch we are 
familiar, and many of its sounds are so singular, — living remnants 
of habits which seem to have been widely diffused in the xth 
century, but which have become lost, and generally misunderstood 
in modern times — that a careful examination and explanation of 
their nature is necessary. As no treatise has as yet appeared which 
conveys satisfactory information, I have availed myself of the kind- 
ness of Mr. Eirikr Mag-nusson,^ who, to a perfect knowledge of his 
native tongue joins a long and familiar acquaintance with the 
language and j)ronunciation of England, and who has taken the 
greatest pains to enable me to render the following account as 
complete and trustworthy as possible.^ Whether the actual pro- 
nunciation of Icelandic is or is not the same as that in use in the 
xth century, it is not easy to determine. The antecedent probability 

and Norway ; then arose the name graphy adopted in ancient manuscripts 
islenska (iis'lenska) which the tongne are given in an appendix. A gram- 
has kept to the present day." — Rask, mar is to follow, and in the meantime, 
Gram. art. 518. "From the North Dasent's Eask's Grammar may he used, 
the same tongue was spread over the The following are Icelandic Diction- 
Ferro, Orkney, Shetland, and Western aries of repute, which have superseded 
Isles, and fi'om Iceland to the coast of Biorn Haldorson's Lexicon Islandico- 
Greenland : hut the old Greenland has Danicum, edited hy Eask, Copenhagen, 
been now for a long time lost, and since 1814, 2 vols., 4to. Sveinbjorn Egils- 
the Scottish Isles were joined to Scot- son, Lexicon Poeticum antiquae Linguae 
land, the Old Norse language has given Septentrionalis, Copenhagen, 1840, 8vo. 
way to the New English. On the pp. 932. Erik Jonsson, Oldnordisk 
Ferro Isles a dialect is still spoken, Ordbog, Copenhagen, 1863, 8vo. Fritz- 
which comes very near to the Icelandic, ner, Ordbog over det gamle norske 
but is of little interest since it has no Sprog, Christiania, 1867. 
literature except some popular son<?s." 9 ttij-. c ix^ • z a-l- t xx^ 

-Ibid. Art. 520. Thesi songs were , ' ^f\f ^^^ ^^I^^f ^ p-n" V t 

IT 1 J i.\^ T\ ■ \ i. 1 J.- V Icelandic Version 01 the iJible lor the 

published with a Danish translation by t> -i- -u x tj^ • -dm i a ■ x. 

V _ 1 -DA looo rT\ J British and Foreio-n Bible Society, 

Lynffbye, Eanders, 1822 (Dasent s ,, o -r i ^ t 1 j J 

note). See also Iv^rAasen's Diction- author of Legends of Iceland, and 

ary of the Dialects of Norway. translator of various sagas. 

1 Prof. Th. Mbbius's Analecta Nor- ^ Mr. Henry Sweet, of the Philolo- 
rcena, and Altnordisches Glossar, re- gical Society, having acquired the pro- 
cently published, will be found useful nunciation of Icelandic from another 
for students who are acquainted with teacher, Mr. Hjaltalin, I requested him 
German. The glossary extends to to inform me where his impressions dif- 
several other selections named in the fered from mine. The observations 
preface. A uniform modern orthogra- which he has been kind enough to 
phy is adopted in all the extracts, but furnish, are added in the shape of foot- 
carefully printed specimens of the ortho- notes, signed H. S. 


is that there are differences, and with respect to y this probability 
amounts almost to a certainty. Eut Eask, Rapp, and Grimm ^ 
differ most materially in their views, and as they cannot all be 
right, it is very likely they are all wrong. N^one of them seem to 
have pursued a satisfactory course for arriving at the truth, which 
would require a long study of the phonetic relations of existing 
dialects in Denmark, Sweden, E^orway, and Iceland, the careful 
examination of ancient manuscripts, of rhymes and assonances, and 
of the internal phonetic relations of the language itself. Mr. Hemy 
Sweet having carried out this programme to a great extent, has 
obligingly furnished me with his own views on the subject, which 
I have appended to a tabular account of the opinions of Rask, 
Eapp, and Grimm, at the close of this section. It is first necessary 
to ascertain existing usage. 

Icelandic now possesses eight simple vowels, «, e, i, i, o, 6, u, u 
= (a, e, i, i, o, oe, 9, u) either short or long, the shortening being 
generally indicated by two following consonants, or a doubled con- 
sonant. The letters y y are at present identical with «, i. It has 
also six diphthongs ; namely, three i diphthongs, m au, ei or ey, 
the two last being at present identical = (aa^, oeoe^, eei) ; two u 
diphthongs, a, 6 = (aau, oou), the great peculiarity of all these 
diphthongs being the importance of the first element, and the 
brevity of the second, which in the case of ei, 6 amounts to that 
faint indication of an (i, u) heard in the English day, know (d<?^'j, 
nooi'w), in Icelandic letters dei, no ; and one acknowledged diph- 
thong with (i) prefixed, e or e as it is now written, and which 
might with equal propriety be written je, for in fact there are 
numerous other diphthongs of the same class, now written with a 
prefixed/, but formerly written with a prefixed i. 

The consonants h, d, h, j, I, m, n, p, r, s, t, v = (b, d, h, j, 1, m, 
n, p, r, s, t, v) almost invariably ; / varies between (f, v) and some- 
times (b, m) ; k, y are properly (k, g) but are often palatalised to 
{k, g), and y takes all guttural phases of (gh, yh, jh ; g^^h, tvh), 
down to (j, w), and complete disappearance ; c used to be employed 
in the combination ck only, and q in the combination qv, but as 
neither c ov q belong to the language, they have been both super- 
seded hj k; ^ is occasionally used for ks, or ys ; and z is employed 
for the sound of s before which a dental has been omitted, but not 
very consistently. The old letters f, ^ are retained as (th, dh), 
although d is often employed for ^ in older printed books. The 
combinations /?/, hi, hi, hr, hv are called (jh, Ih, nh, rh, wh). The 
double letters //, nn are mostly (dl, dn) when medial, and (tlh, tnh) 
or (dtlh, dtnh) when final. In the doubled tt, the first t indicates 
an assimilated guttural, which however is generally more or less 
heard. The following is a particular alphabetical account of the 
behaviour of each letter and principal combination. 

1 A Grammar of the Icelandic or fort, Jaeger, 1843. The Swedish title 

Old Norse Tongue, translated from is : Anvisning till Islandskan eller 

the Swedish of £rasmus Rask by G. Nordiska Fornspraket, af Erasmus 

W.Dasent, London, Pickering; Frank- Christian Eask. FrSn Danskan bfver- 


Icelandic Alphabet. 

A, distinctly (aa, a), not so low as {aa, a), and never rounded to 
(ah), but occasionally as high as (aah, ah), though this may be an 
individual peculiarity, and was certainly unintentional.^ Most of 
the words cited by Grimm as having short (a) are now pronounced 
with long (aa). Ex. hann (nan) he, alt (alht) all, haf^di 
(navdly'), landi^ (landnlli) the land; draii (draavi) htcsks, matar 
(maa'tar) meat = food, taka (taa'ka) tahe, ma^ur (maa'dh^r) mawf 
sag^i (saahLgh-dh/)^ said. In unaccented syllables, where open or 
closed, the short a is general. 

K, a clear diphthong (aau), with the first element predominant, 
and the final short, and thus distinguished from the German au 
(au). Not (ao, ao) as suggested by Rapp. I^ever (aa), but con- 
founded occasionally with o in MSS, with which compare the 
Welch confusion of aw, o (au, oo). When a is final and emphatic 
there seems to be an inclination to sound after it a whispered u 
('u), or the labio-gutturals (wh, g^^h), just slightly touched, as d 
(aaui_wh) river, fa (faau|_wh). Before a doubled letter the first 
element is somewhat shortened, and before doubled t, the guttural 
is decidedly touched, as dtti {aM\kwht-ti) had, but the whole com- 
bination is spoken with extreme brevity. 

JE, the diphthong (aai), taken by Rapp as (ae), from his inability 
to appreciate (?') ; distinct therefore from German ei, at (ai). There 
is an unacknowledged tendency to develop a palato-guttural sound, 
as (j, jh, gh., Jch), after ce, when final, or before a vowel, as : ae 
(aai^jh) a^e ever, aea (aarja) to cnj for pain. And before two con- 
sonants or a doubled consonant, the first element is shortened, as : 
setla (a/t'la) to think settir (a^t•t^r) oiightest. 

ATI sounds to me as the diphthong (ceoe^*), scarcely differing from 
the French ceil on the one hand and the Dutch ui on the other. 
Rask refers the Icelandic sound to the German eu, as Dr. Gehle did 
the Dutch (supra p. 235, n. 1, and p. 295, n. 1), and Rapp, as I 
understand him, says that Rask pronounced the diphthong au as 
(oe^), which pronunciation seems to furnish the key to the ortho- 
graphy, for a changes its sound by U^nlaut to e through a following 
i, and to 6 (oe) through a following ii (9), as : fa^ir, fo^ur (faa•dh^r, 
foeoe'dh^r). This organic law of change was probably the cause 
why au was written for in old MSS. quasi, a as altered by the 
influence of u, and the same spelling was also used for 6u (oeoe^) 
most naturally. JS'ow since {p) is often confounded with (y), and 
(y), when brief, is easily confounded with {i), we see how au might 

satt och oraarbetad af Forfattaren, bles, and in accented intermediate to (ah) 

1818. Physiologie der Sprache von and (a). — H.S. Is this sound (a f-) ? 

Dr. K. M. Rapp, vol. 2 (1839), pp. ^ Compare the Norfolk maivther, a 

128-139, vol. 4 (1841) p. 246. Ver- girl, and the observation in Nail's 

gleichende Grammatik, vol. 3, (1859), Glossary. This Icelandic word was 

pp. 39-41. Deutsche Grammatik von formerly mannr, modern Danish mand. 

Jacob Grimm, vol. 1, 3rd ed., 1840, ^ Pq^. ^j^e use of [ to signify a scarcely 

pp. 421-495, 2nd ed. 1822, pp. 280-330. audible utterance of the following ele- 

Decidedly (ah) in unaccented syUa- ment, see supra, p. 419, note, col. 1. 


come to be (oeoe^, oeoey, oeoe*), and, in the present absence of (y) 
from the language, would naturally rest in (oeoe^). The German 
eu is very variously pronounced (supra p. 321, note 2). Eask must 
have alluded to the somewhat rare (^y) sound, which he heard as 
(oey). If the view here taken be correct, the sound (oe^) was pro- 
bably the oldest form of this diphthong, and the antiquity of the 
{9) sound of u, is also rendered probable.^ Ex. hlaup (Ihoeoeip) 
course, lauf (loeoe«V) leaf, skaut (skoeoe«t) lap, kaupa (kceoerpa) hiiy, 

E is always (b). 

C is ' ' used by old writers indiscriminately with h, especially at 
the end of monosyllables. It is now used only in ch for hk, but 
many write kh and thus shut c entirely out of the language, a 
custom which is already (1818) old, though not general." — Rask. 

D is always intended to be (d) according to the present orthogra- 
phy, but in older printed matter it also stood for ^. It is found only 
at the beginning of words and syllables, and after /, n, m, and d. It is 
occasionally written when not pronounced, as: syndga (sm*ga) to sin. 

D is precisely the English (dh), but never occurs iuitially in 
Icelandic, where it is found in place of (d), after vowels and r, /, g, 
and ''in old writers it is sometimes found after I, w." — Rask. 
There are some districts in Western Iceland where it cannot be 
pronounced, and is replaced by (d). It has disappeared in Swedish, 
but is heard though not written, in Danish. The present use of 
J7, ^ in Icelandic accords generally with their written use in Anglo- 
saxon, and consequently there is a presumption that the English 
use of an initial (dh) is modern, see supra p. 515.^ 

1 This conjecture will be incorrect if, however, maybe a remnant of the form 
as seems probable, Mr. Sweet's views thocht, possibly a form of thought^ for 
are to be adopted, infra, p. 559. which initial (th) would be regular. 

2 Since p. 515 was sent to press, Mr. As regards Anglosaxon, the real usages 
Henry Sweet has read his investigation of MSS., disregarding the manipula- 
of the meaning of )? "S before the Phi- tion of editors, are very uncertain, ac- 
lological Society (4 June, 1869). He cording to Mr. Sweet. The Northum- 
considers that the sound Avas originally brian writings use ^ everywhere, except 
uniformly vocal = (dh), in the earliest in the contraction J^*. Eapp (Yerglei- 
stages of the Teutonic languages, and chende Grammatik, iii, 128) complains 
that the non-vocal (th) is a later and that a great mistake has been made 
progressive development. He believes respecting Anglosaxon ]> "S, especially 
that the earliest Icelandic of the xiii th in England. The Anglosaxons, he 
century had the same pronunciation of says, probably wrote first with runic, 
]? "S as the modern, except in the words then with Latin letters, and there being 
which have exceptionally an initial no Latin letter for (th), the sound was 
vocal form in English, thus, ancient represented in three ways ; occasion- 
^at, '^essi ^w= modern Ipa'^, Ipessi, \u. ally, even in the oldest monuments, by 
But the testimony of Icelandic MSS. th^ [compare supra p. 525, 1. 22] ; 
he finds to be very uncertain. In mo- afterwards by the runic j^, and thirdly 
dern Icelandic, "S is often evanescent by the Icelandic '5. Englishmen could 
(Ldh), according to Mr. Sweet, and in the not but feel that ]>, '(S were convenient 
Norwegian dialects it disappears entirely representatives for their own two sounds 
leaving an hiatus. See Rapp's opinion, (th, dh), although a cursory inspection 
infra p. 555, n., col. 2. It should be men- of the MSS. would shew the discord- 
tioned that one of our words having an ance ; so that some inverted the order 

nitial (dh), though, is pronounced with and made )>, 'S = (dh, th), [supra p. 515, 
initial (th) in Scotland, (thoo), which note 1]. Neither the Anglosaxon nor 


E is properly (ee, e) long and short. ^ The sound did not appear 
to me to be so low as (ee, e), and certainly was not so high as {ee, e). 
Giimm (ib. pp. 427-432) endeavours to divide the sound into 
two, (e) corresponding to Gothic a, and ((?), wliich he writes e, cor- 
responding to Gothic i. There is no trace of this in the spoken 
language. Ex. ennfremur (en"free"mpr) and further ; sem (seem) 
who \ herrar (ner*rar) lords, verk (verk) worlz, etc. Initially it is 
occasionally pronounced like ^, as : eg (jee^li) /. 

E', E', the form h was proposed by Rask, and has been generally 
adopted, the older writers employ e or omit the accent altogether, 
leaving it to be supplied by the reader — either form is considered 
equivalent to je, and should therefore be (jee, je), but in fact, as 
in many cases where j is written, the result is often a diphthong 
with the stress on the first element, as : tre (triee) tree, mer (mi'eer) 
to me ; but : fenu (fjce'uo, fiee^na) fees, ^property, rettur (rjet't^r)'' 
right, fell (fjedtlh) y^//, etc. 

EI, EY. These two signs are now identical in signification. 
Rask says that the two sounds are still distinct in ISTorway, where 
ey = (oei), and in the Ferro dialect, where it is commonly (oi). At 
present, however, both are {ee%) or (<?(?'j), not sensibly differing from 
southern English ^ay, and having its first element distinctly {ee) 
and hence materially differing from e. It is occasionally shortened 
by shortening the first element, and then may be written {e\i') to 
shew the brevity of the second element, so that the effect is almost 
((?). Ex. seil {^eeiX) towing line, heill (H^[_/dtlh) ivhole, j^eirra {\h.e\ix'- 
ra) of them, eytt {eyi'C) wasted. 

E, properly (f), with a very mild hiss, scarcely more than a 
single tooth being touched by the lower lip, so that it approaches 
(ph). It has this sound only at the beginning of syllables, or before 
s. or when doubled. At the end of a word or between vowels it 
falls into an equally mild (v). Before I, n, at the end of syllables 
it falls into (b), but if d or t follow the n, then flid, fnt become 
(mud, mnt), most generally, though some say (mnd, fnt). Ex. 
fotur {ioowioi) foot, ofsi (ovs?) arrogance ; haf (uaav) sea, arfr 
(arv'^r) inheritance; tafia (tab'la) tahle, nafn (nab'nh) name-, nefna 
(neb'na) to name, nefnt (nemnt) supine of nefna ; jafnt (jaft), from 
the pulpit (jamnt) equally.'^ 

G is the most changeable of all the letters, and it is difficult to 
lay down rules which should apply to every case. At the begin- 
ning of syllables it is (g) before a, a, o, 6, u, ii, 6, an, and {g) before 
ce, e, ei, i, i, y, y, ey and also before j. The first group corresponds 

Early English use \ or ^ in place of and 'S eliminated. He even assumes ini- 

an organic (d). The Englishman now tial th — {\X^ in Chaucer, see the intro- 

pronounces the demonstrative pronomi- duction to Chap. YII. \ 1, near the end. 

nal family with initial (dh), which ^ I took the e for (e) instead of (e). 

no one has yet asserted for Anglosaxon — H.S. 

{was noch niemand im Angelsuchsischen ^ The sound before tt is a pure 

behauptet hai). He considers that Eng- aspirate without consonant quality, 

lish (dh) has arisen partly from (th) rett (riEH't). — H.S. 

and partly from (d), and that in Anglo- ^ Jafnt or jamt with voiceless m 

saxon J7, d, must be everywhere restored, (jamht) . — H. S. 


to non-palatal vowels, and the second to palatal vowels, but this 
division is not exact, for e, u (e, d, ce) have precisely the same 
elevation of the tongue as ei {eei), and ce (aai) is a back vowel, 
before which the use of the palatal {g) is exactly similar to that in 
older English regard, sky (ri^aard*, s/(;ai), supra p. 206. The palatal 
Ic, g are expressed by kj, gj before the first group, and should always 
be so expressed. G after «, o, becomes (gh), and after d, u, it falls 
into {wh, wh, w) or almost entirely disappears. Eut after an (i) 
sound, it becomes (^h, /;h) or even completely (jh, j), and occasion- 
ally disappears as (i). These changes are extremely interesting be- 
cause they shew the stages through which the ags. 5 passed in older 
English before it entirely subsided into the present (j i, w u) or 
totally disappeared. We have, therefore, an actual living example 
of the intermediate sounds, already suggested by theory, establish- 
ing the correctness of the previous hypothesis, supra, p. 512. Ex. : 

(g), gafa (gaau'va) gift, gas (gaaus) goose, gaukur (goeoer/t^r) cuckow, 

gl6^ (glooudh) live coal, go^ur (goou dhsir), gora (^oeoe'ra) to 

{g), goes (^aae's) geese, gaeta (_^aarta) to keep, geit (geeit) goat, gjof 

(^^fioeoev) gift, gjarn (^iadtnh) prone, pyngja (piiq*^ia) purse, 

gefa (_^ee*va) give. 
(gh) og (oogh) and, dogum (doeoegh'^m) to days, sag^i (saahi_gh*- 

dh«) daglaunamenn (daai_gh'loeoernamen') day labourers. 

{^ivh., wh, w), Ijuga (ljuu'[_g?i;ha, ljuu"[_wha, Ijuu'wa, ljuu*a) to tell 
a falsehood, all varieties of barely pronounced (g^^h) being per- 
missible, and the last two forms being most common. This 
disappearance of (gw^h) strongly calls to mind the absence of 
(gh) in the Welch system of mutation of initial consonants, 
thus (b, f , m ; d, dh, n) should have in Welch a correspond- 
ing (g, gh, q), but instead of (gh) an hiatus is substituted as : 
eu gafr, dy afr, fy ngafr {ey gaav'r, da aav'r, va:-qaav'r), their, 
thy, my goat, where we ought clearly to have (da ghaav'r). 

(^h, jh) mig (m//^li) me, eigum {eeigh.'dYcC) possessions, sig (sM^h) 
himself, eg (jee^h) /, gnseg^ (g"naa/[_^hd), enough. 

(^h) fjarlaegt (*laa/|_/i;ht) /«r lying. 

(j) feginn (fee'jm) fain, segja (seei'ja) to say, dragi^ (draa*j/dh), 
draw, put, bogi (boo'ji) how for shooting, agi (aa*je) chastisement, 
bagindi (baau•Jmd^) troubles. 

In addition to these we must reckon the cases where a scarcely 
perceptible (^h, jh, ^w\, wh) is developed from [i, uu) as : ae, bu 
(a2|_jh, buu|_wh) ever, farm. The Swedish reading of gn as (qn) is 
unknown except when d, t follow as lygndi (iMqu'd?) became calm, 
rigndi, rignt (rwqn-d?, rwqnt) was rained on. When s follows the 
n is lost, as gagns (gagks). 

H before vowels is (h', h) and is never dropped. Before conso- 
nants it is used simply to make them voiceless. Thus we have the 
remarkable set of digraphs, HJ, HL, HjN", HE, HY, existing as 
distinct (jh, Ih, nh, rh, wh), as was conjectured for Anglosaxon, 


p. 513. HJ = (jh) is precisely the same as the initial element in 
my proniuuiation of hue (jliiu), and is not [Jch., ^h), but of course 
only sliglitly different. HL = (Ih) is the true whispered (1), with 
the breath passing out at each side of the tongue, and hence dif- 
ferent from the unilateral Welch // (Ihh), so that Welch : lladd 
(Ihhaadh) to kill, and Icelandic : hla^ (Ihaadh) a street, a mound, 
are perfectly distinct in sound. This (Ih) sound is also frequently 
developed from II final, intended for dl, but called (dtlh) as all 
(audtlh) eel, and even before t, as : alt (alht) all. It would 
therefore naturally replace our English final ('1) in fiddle, if I 
occurred final after a consonant, just as the modem French stable 
(stablh), p. 52.^ This is really the case withHN = (nh), which not 
only occurs initially, as hnifur (nhii'v^r) hnife,"^ but in w?^ as : einn 
(««dtnh) one, and : vatn (vatnh) water. In HR = (rh) the Icelandic 
possesses perfect whispered r, which on the analogy of (Ih, nh) 
is the sound of the favourite nominative termination -r in old 
Norse, as : bleikr, cleigr {hleeikTh, ^eeigih) pale, ivet, but the modem 
custom is to use -ur (-^r) in its place, and this pronunciation has 
probably arisen from the sound (rh) having been dropped, and (r) 
simply retained, as (bk<?^kr) with a distinct trilled (r) not fonning 
a syllable, and different from (bk^zk'r), into which it probably sank, 
before the transition into (bk^rk^r) took place, as the Icelander 
naturally conceives all indistinct sounds to be [d] which is his 
''natural vowel." The close resemblance of (rh) to (s) however, 
and the correspondence of the Icelandic -r with the Gothic -s, 
renders the old sound (rh) extremely probable, and possibly the 
old Latin confusion of terminal s, r as arhos arhor, honos honor, 
may rest upon a similar antecedent whispered pronunciation of r. 
The use of HY = (wh) is the most singular, because (w) is not a 
recognized element in the language, and it will be best considered 
under Y. 

1 is distinctly {ii, i) both long and short, the very sounds which 
we were led to attribute to i in the xiv th century (p. 297). It is 
interesting also to see that foreigners, unable to appreciate the true 
{ii i), confuse it with {ee, e),^ which is a corroboration of the re- 

^ The sound of hi is more correctly ^ p,ask says that the '' sound espe- 

(Ijh). — H. S. See infra, p. 546, n. 1. cially when it is long seems to approach 

to that of the deep e {e)." Eapp says 

2 Compare Cooper, p. 32, "N For- "folglich i = e gilt," i.e. consequently 
matur ah extremitate linguae superio- i = («). Grimm says: ''"Waihrend der 
rum dentium radici apposita (si spiri- unterschied zwischen i und i in solchen 
tus utrinque per labia efflatur formatur zweisilbigen formen heinahe unmerk- 
l) huic correspondet hn, quam scribunt lich sein, z. b. qvi'Sa poema fast lauten 
Angli per kn^ know know, cognosco." musz wie qvi'Sa metus, obschon kurzes 
• — p. 37, "Aw quam scribimus kn.'" i im munde des Islanders sich dem 
— p. 38, "27«, wh, sh, th, hn in Alpha- elaut nahert," i.e. he considers that the 
beto non numerantur."— p, 39, " kn dissyllables qvi^a poem, qvi'Sa fear 
ponitur pro hn.'' — p. 67. ''JSTwsona-, ought to be nearly indistinguishable, 
tur ut hn ; knave nebulo, knead mala *' although in the mouth of an Icelander 
cisso, A«fe genu, ^;2^<?/ ingeniculor, /c»?2yi? short i approaches to the sound of ^." 
culter, knight eques, knit necto, knock (Gr. P, 486). Mr. Sweet says than 
tundo, know nosco, knuckle articulus ; in unaccented syllables i is rather (e) 
quasi hnave, etc." than [i). 


mark, p. 271, and even in some terminations, e often stands in 
MSS. for «', as in : haskalegr, misseri, lande, for haskaligr, missiri, 
landi (naaus'kalng^r, m«s'snr«, land'w) dangerous, quarter year, to a 
land. At the present day, however, the {ii) is very distinct, as is 
never confused with (ii), thus : vinum minum (vn'n^m mii'n^m) do 
not rhyme, and children in repeating the alphabet never confuse * with 
i, that is {ii) with (ii). Icelandic is the only language I have met 
with which distinctly recognizes this long (n ), though we have seen 
that it is occasionally generated in English (p. 106). The short i 
is the true usual English («'), and is perfectly distinct from (i). In 
older books i before a vowel was used, where/ is now employed. 

r on the other hand is (ii, i), generally long, but short in un- 
accented syllables. It is not, however, found short in closed ac- 
cented syllables as in Scotch and French.^ Rask considers z, u as 
diphthongs, as it were ij uv = {ii, 9u), but there is no foundation for 
this in actual speech, and the conception seems due to the mode of 

J was used as the ancient capital of i, at the beginning of words, 
but as it was there pronounced as (j) before vowels, it has in recent 
times been used in the middle of words before vowels, even though 
the sound was not always the pure consonant (j), but much more 
frequently an (i) diphthongising with the following vowel. It 
changes a preceding k, g from (k, g) into (Jc, g), but the sound of (i) 
is still heard as much as in the Italian : chiaro, ghiaja (/^iaa*ro, 
^iai'ia) clear, gravel. It does not seem to change a preceding I, n 
from (1, n) into (Ij, nj), as Ija (liaau) new cut grass, lj6^ (lidoudh) 
poem, liufur (liuu'v^r) gentle, Ijae (liaa^) to lend; nialgur (niaaul-- 
g^r) hedgehog . In some cases the sound of (j) would be difficult 
as : fjarins (fiaau'rinzs) of the fee, fjarlaegt (fiaarlaa^L^-ht) far-lying, 
bjost (biooust) husked, brjosti (briooust*/) hreast, hljop (Ihiooup) 
leaped. Hence/ must be merely looked upon as a dipthongizing (i), 
not {i). In all these cases, however, a simple (j) would be con- 
sidered correct, thus (Ijaau, Ijooudh, ljuuv',?r, 'Ijaae, njaaul'g^r, 
fjaau'rins, fjaar"laa«[_i?;ht, bjooust, brjoous't?', Ihjooup). 

K is (k) before a, a, o, 6, u, ii, d, au and {Ji) before (2, e, ei, i, i, 
y, y, ey, j, thus kirkja (hirkidi) church, contains the true inter- 
mediate sound between the Scotch kirk (kerk) and Chaucer's chirche 
(tshirtsh'e), supra pp. 203-6.^ K does not assume the forms (kh, 
^h, \w\i), and hence differs materially from Gr. 

L is usually and always intentionally (1), but the sound of (Ih) 
is sometimes produced by a following t, as alt (alht) all. In the 
case of II, the first I is pronounced as (d), and if the second is final, 
it becomes Qh), and thus generates a (t) in passing from (d), so 
that the combination becomes (-dtlh), and the first (d) is frequently 
scarcely audible, as (-i_dtlh), the whole combination being rapidly 

1 Short (i) in ]?ing (thiqg). I think ^ j thought k before e, i, etc., was 

Mr. Hjaltalin said that the pronuncia- really (kj) not (kj = Ar), hut this was 

tion (thqg) with open (i) sometimes probably incorrect. — H. S. 
occurs. — H. S. 



pronounced/ and rl is treated in the same way, thus : kail karl 
(kaKltlh) calling, churl. Between two vowels, II is distinctly (dl) 
as kalla (kad'la) to call. See N. For hi see H. 

M is always intentionally (m), but may be voiceless (mh) before t. 

1^ is always intentionally (n), but after t, k final, (nh) is generated 
as : vatn (vatnh) water, regn (reg|_knh) rain, vagn (vag|_knh) wain^ 
and mi rn are both (-i_dtnli) final, see L. Thus klenn (klicLdtnh) 
small, finn (fii^dtuh) jine, jam (jaau|_dtnh) iron. " But should 
nn belong to the following syllable, or if it be a simple vowel that 
goes before, the sound is (n), as a-nni (aun'm) to the river, dat. sing, 
with art., ey-nni (^m*m) to the island y' so also : kanna (kan*na) to 
survey, hann (nan) he, brenna (bren*na) to hum, etc. Old writers 
often used //, nn, in all cases before d, t without regard to the 
radical form, though the custom was never general. This nnd has 
been long since entirely laid aside, as also II, d where the root has 
a simple /, //"." — Bask.^ In iN'G the n becomes (q), and the g has 
its full sound of (g), thus fing (thiiqg)* council, assemhly, and the 
preceding vowel is always one of the accented series a, i, 6, u, y. 
Konra^ Gislason, however, maintains that the vowel should always 
be unaccented in old Norse ; but his opinion does not find much 
favour. ]S^K is also pronounced (qk) as : j^anki (thaauqk•^) mind^ 
thought, hanki (naanqk**) handle of a basket, ear of a jar. 

is the pure (oo,) long and short, supra pp. 94-96, quite dis- 
tinct from the English (aa, o),^ and is identified by Bask with the 
Swedish a, Bussian and Finnish o, but as he also makes it the same 
as English o (o), some doubt attaches to the other indications. 

0' is the pure English diphthong (oou) as heard in know. The 
final u here generates a (w) when another vowel follows, as soa^i 
(soo'wadhi) wasted. When a doubled tt follows, where there is an 
assimilated guttural, the first element is shortened, and the guttural 
is faintly heard, as dottir (dou|_g«^ht't«r) daughter. When 6 is final, 
the (u) is heard quite as distinctly as in EngUsh, thus sko (skoou) 
shoes, is a perfect rhyme to know. 

0, (E,^ is (oeoe, oe) long and short, and is kept quite distinct from 
(99, d), as in dogum (doeoegh'g'm) to days. The form oe is only used 
by theoretical writers. 

B is always (p), except in the combination 'pt which is called (ft) 
as lopt (loft) air, but modem writers, and among them the learned 
Jon porkelsson, are beginning to employ /(( by preference. 

^ Z/, nn = ('dlh, 'dnh) between bias being etymological, not phonetic , 

vowels generally, as well as final, falla, in Icelandic, -nn is said to belong en- 

allra, ^mj^za^ = (a'dlhrah), etc. L is tirely to the second syllable, but a dis- 

generally rather (Ij). One Icelander tinct (n,n) is really pronounced. 
(Mr. G. Vigfusson) said he could not ^ Before t, n is voiceless as heint 

sound the English /. Thus/a^/a is more (bmnht). — H.S. 
correctly (fa'dljhah). — H.S. * See p. 545, note 1. 

2 In both these cases -nni stands for ^ I took the o for (o) not (o). — H.S. 

-imii and is the dat. fern, of the suffixed ^ In old Icelandic there was a long 

definite article, so that it has no ety- oe distinct from se, but it seems to have 

mological connection with the preced- been absorbed by se at an early period, 

ing d, ey, and the division of the sylla- — H.S. 


Q,Y is found in old MSS. but even there interchanges witli lev. 
At present q has no value dijfferent from (k), and consequently (Jc) 
is now generally written. 

E is a strongly trilled (.r) as in Scotland, and when doubled, as 
in fjarri (fia.r'.r^) remote, the number of vibrations of the tip of the 
tongue is very great. Final -ur {-9t) is however more lightly pro- 
nounced. In the following transcription I shall simply use (r). 
but the reader must be careful never to say (j). The combinations 
rl, rn are considered under L, N". The final -r after consonants, 
was probably (rh) see hr under H, but it is now generally replaced 
by -ur {-9r)} 

S is always intentionally (s), and never (z), but (z) is sometimes 
generated, although it is not recognized. Thus (s) final after I, n, 
and perhaps in other cases, generates an intermediate (z). For ex- 
ample, if we compare : eins, sins {eeinzs, siinzs), with English 
stains, scenes (st^einzs, siinzs), we shall see that the difference 
of the terminations, here written alike, arises from the (s) in Ice- 
landic being intentional and predominant, but the (z) generated 
and therefore lightly touched, while in English the (z) is inten- 
tional and predominant, and though the (s) is often prolonged, 
and in the church singing of charity children, not unfrequently 
painfully hissed, it is yet merely generated by a careless relaxation 
of the voice, and its very existence is unknown to many speakers. 
We might therefore write the Icelandic (-n[_zs) and the English 
(-nz^s), but (-ns, -nz) is sufficient for most purposes. I found also 
that there was an unacknowledged tendency to pronounce s final 
after long vowels, in the same way ; thus : las, bas, meis, vis, hris, 
ros, hus, mus sounded to me (laauzs, baauzs, m^^«'zs, viizs, rhiizs, 
roouzs, Huuzs, muuzs) halter, stable, maiiger, wise, vegetable, rose, 
Jiouse, mouse, the two last words sounding quite different from the 
Scotch (hus, mus). Even in the name of Iceland itself, I'sland, I found 
the s varying from (z) to (s) at different times, as (iis'land, iiz'land). 
Between two vowels s may similarly have a tendency to become (z), 
but I have not had time to examine the numerous words of this 
class orally, and it would be necessary to examine natives who had 
not learned the sound of (z) from other languages. We may 
always pronounce (s) without offence, but (z) would be frequently 
very off'ensive. Initially before j, s seems to assume the form (sj) 
or (shj), the latter was the sound I heard in sjukur (shjuu'k^r) 
sich. Icelanders have a difficulty in acquiring the sound of English 
(sh), except in such a word as sugar, which they probably call 

T is the usual (t), but in tt, where the first t stands for an assimi- 
lated guttural, while both letters are pronounced (t,t), the guttural 
still generally asserts itself, see ^, K, 0'. 

p is (th), and that invariably, although it stands in places where 

^ In rt, the r is voiceless, as hart pronounce (sh, tsh). They sound our 
(narht). — H.S. cAe^rcA as (siErhs). They also find our 

"^ Most Icelanders seem unable to (z) very diflGlcult. — H.S. 


(dh) is now pronounced in English. Rask, however, excepts ** pro- 
nouns and particles which in daily speech are attracted like en- 
clitics to the foregoing word, as a acfi-finni' in thy days, haiir ]7u 
hast thou ? where it has the sound of ^. The word J7u is often thus 
contracted with verbs, in which case u loses its accent, and } is 
changed into ^, dy or ^, as the foregoing letter may require ; as 
haf-^u (navdh^) Imper. of hafa to have, koni-du (kom"d^) Imper. 
of koma, ris-tu (riis't^) of risa to riseV These are equivalent to 
Chaucer's saystow wiltow (sais'tu, wilt'u), sayest thou, wilt thoUy 
(supra p. 371, art. 98, c, Ex.) the vulgar Geiman haschte (nash't?) 
= hast du, hast thou, etc. They are generated, unintentional sounds.* 

U seems to be pure {99, 9) long and short, and the existence of 
the forms a, 6 (aau, oou) would seem to indicate the absence of any 
letter for (u) even in ancient times, and au for (oeoe) and (ceoe^) ap- 
pears to imply that this value of u was ancient, see AU.^ This 
sound of (9) is often confused with (y), on the one hand, and (oe) on 
the other. Thus to Mr. M. Bell the French u sounds {9), and to 
me (y). In our own provinces (y, 9) seem to be heard indifferently, 
thus I heard both (tyy) and {t99) for two in Norwich. See also the 
Devonshire sounds in (p. 301 note). In Scotland (y) and {9) are 
both used, though only (y) is generally recognized. I hear {9) for 
the French e muet, but others hear (a, ^h). In some parts of Ger- 
many (oe) and in others {9) are used for 6. Hence we must not be 
surprised at Rask's finding Icelandic u *' almost like deep* Swedish 6 
in hoy roX*," probably (^), or ''German ii," which he may have heard as 
{9), wishing to keep it distinct from (i) into which his own Danish 
y had fallen. He adds that "the word gu^ God is pronounced 
nearly as gvo^ or gvii^," but to me it sounds (gv^^dh) or {gw99dh) 
where the inserted v, or a labialized y arising perhaps from an 
intense effort to avoid any palatisation of the y into (gj). The 
distinction between the sounds of u, 6 {9, oe) is, if I rightly ap- 
preciate it, precisely the same as that between i, i (i, i), or \e, e) 
that is, the position of the tongue and lips is the same for both 
elements in each pair, but the whole of the back part of the mouth 
etc., is wider for the second element in each pair than for the first. 

IT is (uu, u), long in accented, short in open unaccented syllables.^ 
Eask says that it has two sounds, apparently (uu, u), but his expla- 
nation is quite unintelligible, owing to his confusing vowels so un- 
like, as (o, ^, 0, u). 'No such distinction was admitted by Mr. 
Magnusson. It seems impossible to an Icelander to pronounce 
final u without some labio-guttural intonation after it, such as (wh, 
gwh), thus : bu (buu) or rather (buu|_wh) farm. 

* The change of ]? to "S is rare in ology, thus art. 15, he speaks of " a or 

this case. high e m the Swedish word engel, 

2 See note on ^, supra, p. 541, n. 2. French e in apres, English e in fellow 

^ See, however, a different opinion or ai in hair,^' and "the lower sound 

advanced by Mr. Sweet, infra, p. 559. of e in the Swedish lefva, veta, French 


1 Rask calls (e) deep, and (e) high, ^ Short (u) in ]?6ngr, not {9) as if 

which is contrary to the usual termin- spelled u. — H.S. 


Y is (v) with so slight a contact of the lower lips with the upper 
teeth as to vary in effect at different times as (bh, v), but I did not 
feel justified in noting it as (bh) without having an opportunity of 
hearing the sound from numerous speakers.-^ That it was not 
originally (v) is clear to me from the combination HV, which 
is called (wh) in the southern, and (k^^^h) in the northern districts 
of Iceland, corresponding to the English and Scotch sounds of 
why and the South and J^orth "Wales pronunciation of chw. These 
point to an original (w) and to the transitional sound (bh) before 
falling into (v). For the unvoiced (v) could only be (f), the 
Aberdeen expression of wh ; and the unvoiced (bh) would be 
(ph), neither of which sounds seem to be used, although / now 
falls into v. It is very possible that in earlier times / had the 
true sounds of (f, v), and that «?, then not distinguished in writing 
from u, was (w), whence hv would be (wh). At the present day, v, 
hv = (v, wh) is an anomaly, which could hardly have been original. 

X is traditionally used for hs, gs, without any known reason, 
except custom, and shortens the preceding vowel like a doubled 

Y has precisely the same value as i {i) and is only employed to 
point out certain grammatical or etymological relations. Eut in 
some valleys it is yet called (y), and this was possibly its original 
sound. The present sound is supposed to have taken its rise in the 
XII th century, and to have become prevalent in the xiv th. 

Y' is now the same as z (ii). '' The name of the letter, however, 
is pronounced altogether as it is in Swedish and Danish," says 
E,ask, that is, as (yy) or more commonly ypsilon. 

Z has always the sound of (s), its use is merely etymological 
or literary, shewing that some letter has been lost before s, and as 
it is not consistently employed, it would be better disused altogether. 

The alphabet is read thus, in Icelandic orthography ; a a be ce 
de e^ e e eff ge ha i i jo^ ka ell emm enn o 6 p^ qu err ess te u u 
vaff ex ypsilon ypsilon zeta ]7orn se = (aa aau bjee sjee djee 
eedh ee jee ef ^jee naau ii ii joodh kaau edtlh em en oo oou pjee 
kuu er es tjee 99 uu vaf eks «ps^lon iip•s^lon see'ta thodtnh aa«). 
Eoth 88 and oe are written occasionally, but they are not distin- 
guished in sound, and are both named (aa^'). 

The stress is on the first syllable of all words long or short, 
simple or compound, but in the case of compounds each component 
has an accent as if it were simple, and the chief stress lies on the 
first. A single fijial consonant, or a single consonant between two 
vowels, leaves the preceding vowel long, as : vel (veel) well, man- 
saugur (maan'soeoei'j^r) lovesong, ve^ (veedh) pledge, ]7at (thaat) 
that, til (twl) to. A doubled consonant, or two consonants (of 
which final r is not one) shortens and " stops" the preceding vowel, 
and diminishes the length of the first element of diphthongs. 
Doubled consonants are fully pronounced, as in Italian, supra p. 55. 

1 I thought at first that v was (bh), Mr. Hjaltalin that it was a dental 
and I was only induced to consider it sound. — H. S. 
as a (v) by the distinct statement of 


E,ask asserts that all vowels and diphthongs are nasalized when 
standing immediately before m and w, but if such nasalisation 
exists, it must be very sKght, and I did not detect it. But see 
infra p. 558, 1. 25. 

When three consonants come together one is usually omitted, as 
halft (naaulht) half, volgt (volht) lukewarm^ margt (maart) much. 
Similarly islenzskt (iis'lenst) Icelandic, danskt (danst) Banish ; 
gagns (gagks) of use, hi-afns (rhafs) a crowds, vatns (vas) water's. 
Similarly r is little heard before st and nd, as verstur (vest'^r) worst, 
fyrstur {fis'toT) first. For /-/, rn, see L,N; for fnd, fnt, see F, for 
gnd, gnt, see G. 

These observations will give the reader a tolerably complete 
notion of Icelandic pronunciation, and enable him, with a little 
attention, to read intelligibly. There is no sound really difficult in 
the language, but the combinations are unusual, and will require 
care. It is therefore necessary to have an example, for which, as 
already mentioned (p. 534,) the parable of the Prodigal Son has 
been selected. The text is taken from that revised by Mr. Magnus- 
son,^ and the pronunciation was written down from his dictation, 
and afterwards carefully compared with his reading. The transla- 
tion is constructed on the same principles as before (p. 534). The 
reader is recommended to read the words of one verse over with 
care and repeat them till he can form the sounds with ease and 
rapidity from memory before proceeding to a second verse. If he 
proceeds through the whole parable in this way, and commits the 
text to memory, he will be able to read any Icelandic book in- 
telligibly to an Icelander. 

Lukasar Gu^spjall 15, 11-32. Luuk'asar Gv^^dh'spiatlh, 15, 


11. Ennfremur sag^i hann: 11. En'free'm^r saa[_gh'dhe 
ma^ur nokkur atti tvo sonu, Han : maa'dh^r nok'k^r aul_k^(;ht•- 

t^ tvoo soo'n^, 

12. Sa yngri J^eirra sag^i vi^ 12. Saauiiq*gr«th^[«r*ra saa^ghl 
fo^ur sinn : fa^ir ! lat mig fa dh^' v^'dh foeoedh'^r sm : faa•dh^r ! 
fann hluta fjarins, sem mer laaut vniigh faau than Ih^^'ta 
ber; og hann skipti milli feirra fmau'rinzs, seem mieer beer; 
fenu. oogh nan Bhii'ii m^d'k' fhe\_inQ. 


Verbatim Translation. 
Luke's Gospel, 15, 11-32. father his: father! let me fang that 

11. Still-fiirther said he : man cer- lot of-the-fee which to-me are-borne ; 
tain had two sons ^-nd he divided between them fee-the. 

12. The younger of -them said to 13. Some days since, took the 

^ Hi's Nya Testamenti Drottins vors The New Testament of-Lord ours Jesus 

Jesu Krists, ^samt me'S Davi'Ss Sal- Christ, together with Davids Psalms, 

mum. Endursko^uS utgafa. Oxford : Eevised Edition. Oxford ; printed in 

prenta^ i Prentsmi'Sju Haskolans i Print-smithy of - High - school - the in 

Oxford, k Kostna'S hius Brezka og Oxford, at cost of- the British and 

Erlenda Bifliufelags. 1863. Literally ; Foreign Bible-fellowship. 

§ 4, No. 2. 



13. N^okkrum dogum si^ar tok 
sa yngri alt fe sitt og fer^a^ist 
1 fjarlaegt land ; far soa^i hann 
fe sinu 1 ohofsomum lifaa^i. 

14. !N"u er hann haf^i eytt 
oUum eigum sinum, kom J'ar 
miki^ hallaeri i landi^, tok 
hann fa a^ li^a nau^, 

15. Por hann fa og re^st til 
eins borgara i fvi landi, sem 
sendi hann ut a bii sitt, a^ gaeta 
far svina sinna ; 

16. Yar^ hann fa feginn, a^ 
se^ja sig af drafi fvi, er svmin 
atu ; og einginn yar^ til a^ 
gefa honum nokku^. 

17. ^u er hann ranka^i vi^ 
ser, sag^i hann: hversu marga 
daglaunamenn heldur fa^ir 
minn, sem hafa gnseg^ matar en 
eg ferst i hungri ; 

18. Eg vil taka mig upp og 
fara til fo^ur mins, og segja vi^ 
hann : Fa^ir ! eg hefi syndga^ 
moti himninum og fyrir f er, 

19. Og er ekki leingur ver^ur 
a^ heita sonur finn. Far fu 
med mig eins og einn af dag- 
launamonnum f inum. 

13. N'ok'kr^m doeoeghvm sii*- 
dhar toouk saau iiq•gr^ alht fjee 
set oogh fer'dhadhest ii fiaar*- 
laa«|_^ht land; thaar soo'wadht 
Han free sii'no ii oowKoouy- 
soeoem^m U'b'nadhe. 

14. IS^uu er Han navdhe e[_it 
cet'l^m eei\_ffh9m. siin'^m, koom 
thaar jRirkidh. nad'laam ii land'- 
«dh, toouknan thaau aadh lii'dha 

15. 'FoouY nan thaau oogh 
rieedhst tiil mnzs bor-gara ii 
thvii land'e, seem send'^'nan uut 
aau buu|_gz^h sit, aadh ^aaz't'a 
thar sviin-a sm*na. 

16. Yardh nan thaau feevm, 
aadh seedh'ja sitgh. aav draave 
thvii, er sviin'm aau't^, oogh 
^|_/q*gm vardh tiil aadh ^ee'va 
Hoo'n^m nok'kg'dh. 

17. IS^uu er nan rauq'kadhi 
veVdh sieer, saahLgh'dhe nan : 
wher*s9 marg'a daa|_ghioeoerna- 
men* neld'^r faa*dher mm, seem 
Haava gnaa«"L^hdh maa'tar en 
jee^h ferst ii nuuq'grj 

18. Jee_^h yH taa*ka mn^h s'p, 
oogh faa'ra t«Vl foeoedh'^r miinzs, 
oogh seei'ja v«"dh Han : Faadh'er! 
jee^h Heeve sm-gadh troowU 
Hm'nm^m oogh fw"r/r thieer, 

19. Oogh er ek'ki leeiiq'g9r 
verdh'^r aadh iLeei'ta soo-n^r thm. 
Faar thuu meedh mn'i_gh eeinzs 
oogh 6^itnh av daaLghdoeoerna- 
moen'n^m thiin*^m. 

Verbatim Translation. 

younger all fee his and fared in far- 
lying land ; there wasted he fee his in 
un-measure-some living. 

14. Now as he had wasted all own- 
ings his, came there much hard-ear- 
ing (famine) in land-the, took he then 
to suflFer need. 

15. Fared he then and betook-him 
to one citizen in that land, who sent 
him out to bigging (farm) his, to keep 
there swine his : 

16. "Was he then fain to fill himself 
of husks those, which swine-the ate; 

and no-one worth to (became to, was 
at hand) to give him anything. 

17. Now, as he came to himself, said 
he : how many day-loans-men holds 
father mine, who have enough meat 
and 1 perish in hunger ; 

18. I will take me up and fare til 
father mine, and say to him : Father ! 
I have sinned against heaven-the and 
before thee, 

19. And am not longer worthy to 
hight son thine. Fare thou with me 
like as one of day-loans-men thine. 



Chap. V. 

20. Bjost hann fa til fer^ar 
til fu^ur sins ; en er hann var 
enn nu langt { burtu, sa fa^ir 
hans hann og kendi i brjosti urn 
hann, hljop og fell um hals 
honum og kysti hann. 

21. En soniirinn sag'^i vi^ 
hann : Fa^ir minn, eg hefi synd- 
ga^ moti himninum og fyrir 
fer, og er nu ekki framar ver^ur 
a^ heita somir ]?imi. 

22. pa sag^i fa^irinn vi^ 
]7J6na si'na: fa3ri^ hi'nga^ hina 
heztu skikkju og faerie hann i ; 
dragi^ hring a hond hans og 
sko a faetur honum ; 

23. Komi^ me^ alikalf og 
slatri^, svo ver getum matazt 
og vcri^ gla^ir ; 

24. pvf ]7essi sonur minn, sem 
var dau^ur, er lifna^ur aptur, 
og hann, sem tyndur var er 
fundinn ; toku menn nu a^ 

25. En svo bar vi^, a^ eldri 
bro^ir hans var a akri, og er 
hann kom og nalga^ist husi^, 
heyr^i hann samsaung og dans ; 

26. Kalla^i hann ]7a a einn 
af ]7J onustumonnunum, og fretti 
hann, hva^ um vaeri ; 

20. Biooust Han thaau tul 
ferdh'ar tiil foeoedh'^r siinzs ; en 
er nan vaar en nuu laauqt ii 
bM'^, saau faadh'zr nans nan 
oogh X-end"* ii briooust'i ^m nan, 
Ihiooup oogh fi'etlh ?m naaulzs 
Hoo'n^m oogh kts'ti nan. 

21. En soo-Ui^rin saah^gh dh* 
vwdh nan : Eaadh'er mm, jee^h 
neevi sm*gadh moou't* uirn'm- 
n^m oogh Ui'iir thieer, oogh er 
nuu ek'^i fraa*mar verdh'^r aadh 
ne^fta soo'n^r thm. 

22. Thaau saah^gh-dhi faadh*- 
trm viVdh thioo'na sii'na ; faai*- 
redh niiq'gadh n/rna best"^ 
Bki'k'h oogh faa/r"«dh nan ii ; 
di^aau'JKlh rhiiqgaau noendnanzs 
oogh skoou aau faaz'tvr noo'n^m. 

23. Koom'zdh meedh aa'li- 
kaaulv, oogh slaau*tr«dh, svoo 
vieer ^ect'^m maa'tast oogh 
vee*r/dh giaa*dh«r ; 

24. Thvii thes'si soo'n^r mm 
seem vaar dceoeidh'^'r, er lib*- 
nadh^r aft'^r, oogh nan seem 
tiin'd^r vaar, er fend'm ; toouk*^ 
men nuu adh gleedh'jast. 

25. En svoo baar vzVdh, aadh 
el'dre brooudh'«r nanzs vaar aau 
aa'kre, oogh er nan koom oogh 
naaurgadhtst Huus'«dh, nmr'- 
dhi nan saam'soeoe/q oogh dans ; 

26. Kadiadh^ nan thaau aau 
mtLnh av thioou'n^st^'moen*- 
119119m, oogh friet't* nan, whaadh 
^m vaarri ; 

Verbatim Translation. 

20. Busked (arose) lie then to faring 
to father his ; but as he was even now 
long on way (away), saw father his him 
and moved in breast for him, leaped 
and fell over neck to-him and kissed 

21. But son-the said to him : Father 
mine, I have sinned against heaven- 
the and before thee, and am now not 
further worthy to hight son thine. 

22. Then said father-the to thanes 
his : Fare hither the best robe and 
fare him in ; drag ring on hand his 
and shoes on feet to-him. 

23. Come with fatted -calf and 
slaughter, so we get to-eat and be glad ; 

24. For this son mine who was dead, 
is enlivened again, and he, who tined 
(lost) was, is found. Took men now 
to gladden -them selves. 

25. But so bore to, that elder brother 
his was on acre, and as he came and 
neared house-the, heard he music and 
dance ; 

26. Called he then on one of thanes- 
men-the, and asked him, what about 

§ 4, No. 2. 



27. Hann sag^i : bro^ir finn 
er kominn, og fa^ir j^inn hefir 
slatra^ alikalfi, af fvi hann 
heimti son sinn heilan heim. 

28. Eeiddist hann fa og vildi 
ekki fara inn. Fa^ir hans for 
]>y{ ut og bau^ honuni inn a^ 

29. En hann svara^i og sag^i 
vi^ fo^ur sinn : i svo morg ar 
hefi. eg nu J^jona^ J^er og aldrei 
breytt ut af bo^um ]?inum, Jjo 
hefir ]7U aldrei gefi^ mer killing, 
svo a^ eg gaeti glatt mig nie% 
vinum minum : 

30. En ]7essi sonur finn, sem 
soa^ hefir eigum J^mum me^ 
skaekjum, er nu kominn, og 
hans vegna slatrar j^u alikalfii. 

31. En hann sag^i vi^ hann : 
sonur minn, J)u ert alt af me^ 
mer, og allar minar eigur heyra 
]>er til ; 

32. 'Nu settir ]?u a^ vera 
giaour og i go^u skapi, ]7ar 
bro^ir J^inn, sem dau^ur var, 
er lifna^ur aptur, og hann, sem 
tyndur var, er fundinn. 

27. Han saah[_gh'dhe : 
brooudh'«r thm er koom'm, oogh 
faadh'^r thm Heev«r slaaut'radh 
aa•l^kaaul•v^, av thvii nan 
Hmm*t« soon sm JLeei'lan nmm, 

28. R^/d*d?'st nan thaau, oogh 
Ytl'di ekki faa'ra m. raadh*«r 
Hanzs foour thvii nut, oogh 
boeoe^dh noo'n^m m aadh koom'a. 

29. En Han svaa'radh* oogh 
saah[_gh•dh^ vwdh foeoedh'^r sm : 
ii svoo moerg aaur Heev« jee^h 
nuu thioou'nadh thieer oogh 
al'dr^^i hreit uut av boodh'^m 
thiin'^m, thoou neev/r thuu 
sl'dreei gee'vi'dih mieer* k/dh'Uq 
svoo adh jee^li ^aart« glat iRugh. 
meedh vwn'^m miin'^m ; 

30. En thes'Si soom^r thm, 
seem soo'wadh Heev^'r eeiffh'om. 
thiim^m meedh skaai'/cj^m., er 
nuu koom'/n, oogh nanzs veg'na 
slaau'trar thuu aa'likaaul've. 

31. En Han saahLgh'dh* vndh 
Han : soo*n^r mm, thuu ert alht 
av meedh mieer, oogh adt'lar 
miin'ar eeig\\.'9i: Tieei'Tdi thieer till; 

32. ISTuu ai[_/;ht•t^r thuu aadh 
veer'a giaadh'^r oogh ii gooudh*? 
skaa'p«, thaar brooudh•^r thm, 
seem doeoe/dh"^r vaar, er b'b*- 
nadh^r aft'^r, oogh nan, seem 
tiind'^r vaar, er f^nd'm. 

Verbatim Translation. 

27. He said : Brother thine is come, 
and father thine has slaughtered fatted- 
calf, for that he fetched son his whole 

28. Grew-wroth he then and would 
not fare in. Father his fared then out 
and bade him in to come. 

29. But he answered and said to 
father his : In so many years have I 
now thaned (served) thee and never 
deviated out of biddings thine, though 
hast thou never given me kid, so that 
I might gladden myself with fiiends 

30. But this son thine, who wasted 
has ownings thine with harlots, is 
now come, and his ways (for his sake) 
slaughtered thou fatted-calf. 

31. But he said to him : Son mine, 
thou art all of (always) with me, and 
^1 my ownings belong thee to : 

32. Now oughtest thou to be glad 
and in good shape, there (because) 
brother thine who dead was, is en- 
livened again, and he, who tined was, 
is found. 



Chap. V. 

Pronunciation of Old Norse. 

Eask considers that the modem pronunciation is practically the 
same as the ancient, except in a few instances, hence in the follow- 
ing table the modem forms as already explained, are given in Rask's 
column, and his supposed ancient values are bracketed. Rapp gives 
an opinion upon nearly every letter in the alphabet, and although he 
did not consider that he had arrived at a result sufficiently definite 
to give an example, he has transcribed a large number of words into 
his alphabet, a selection of which is subjoined. Grimm's pronun- 
ciation is not easy to be determined, and the sounds which I have 
given must be therefore considered to be in great part conjectural. 
The vowels are taken from the third, and the consonants from the 
second edition of his Grammar. 

On these conjectures generally I make no observation, except 
to remark that I feel doubtful as to the value which Rask meant to 
ascribe to the old u. He says: ^'w, without accent, may perhaps 
have had the sound of the short English u in nut, hut, the Danish 
in hos, the Swedish o in sporde, mennishor Idrorih, etc." These 
sounds are certainly not identical, and I have been accustomed to 
consider them as (o, o, ti) respectively. Grimm assumes the Eng- 
lish u to be a sound between German o and o, whatever that may 
mean.^ IS^either he nor Eask, therefore, had mastered the English 
(a, a) sounds. I have represented Eask's ancient u by {p, u) doubt- 
fully, but believe that the latter is more probable. 


Modern & 





Modern & 




aa, a 




k, k 

k, kj 



















ceoe* foeul 








d, dh 

d, dh 

d, th 





q. qg 










ee, e 

e e 


oeoe, oe 



e, e 

jee, je 

ee, jee 

ee, e 


(not used) 












eei [cei] 





r, rh 



^ { 


g» 9, gb 



g» gj 











99, 9 \0, U\ 


oe, u 

























ii, i [9] 




tV, i 




ii [nj 














1 Gr. P, 391, "vor einfachen con- 
sonanten hat u einen laut zwischen 

nhd. und o ; das nnl. u neigt sich 
mehr zu ii." 

$ 4, No. 2. 



Old ISTorse words as pronounced by Rapp : a (aa) in, water, se (ee) 
always, atta (AAt'ta) eight, auk (ouk) <?/sci, auga (oug-a) eye, bleikr 
(bleik'r) pale, bleydi (bloydli'i) fear, bles (blees) blew, blod (blood) 
hlood, bokr(boeoek'r) looks, bok (book) hooh, bruda (bruudh'a) ofhrides, 
byd (byydh) invite, byggia (byg'ja) huild, dagr degi dogum (dag'r, 
d(9g*i, d^g'um) day, to a day, to days, dottir (doot'tir) daughter, dypi 
(dyyp'i) depth, ey (ay) island, eyk (oik) oah, fel (fjel) fell, fliuga 
(fliu*ga)^y, fotr (foeoet'r) feet, fri (frii)/r^^, fullr (fodl'r) full, fylli 
(fyl'i) fullness, gees (^ees) geese, gas (gAAs) goose, gora gjort {gdn-B. 
gjoort) ^0 ^0, c?^¥, balmr (khAAlm'r) halm, blaup (khbup) leap, hniosa 
(khnioo'sa) sneeze, lireinn (khroin'n) pure, bvitr (khbhiitT) white, 
kaupa (koup'a) to huy, kne (kn^^) knee, krankr (krAAqk'r) sick, 
liuga (liuga) to tell a falsehood, opt (opt) often, skapt (skapt) handle 
ungr (uuq'g'r) youth, verd (bherd) price, vis (bhiis) ivise} 

The following observations on the Old N^orse pronunciation, based 
upon a phonetic examination of the structure of the language, its 
connection with the Teutonic branches and the usages of Old MSS., 
are drawn up from notes kindly furnished me by Mr. Heniy Sweet, 
of the Philological Society (supra p. 539, 1. 9). 

^ The following is a translation of 
Dr. Rapp's latest views on tke subject 
(Vergl. Gramm. iii. 40). "Of the 
seven long vowels, the two strongest 
(ii) and (uu) have remained intact. 
The (aa) subsequently, as everywhere 
else, degenerated in the direction of (o) . 
The mutates of (aa, uu) must here be 
(ee, yy). There must be an (ou) cor- 
responding to the old German diph- 
thong ei, but it is here written au, since 
the mutate, if written ey, could only 
mean (oy) ; the Norwegian dialects re- 
tain (ou). Long (oo) afterwards be- 
came diphthongal, and its mutate coin- 
cides with ce (ee). The third long 
vowel wavers between gothic (iu), be- 
coming, when softened [gescJnvdcht) 
(io), and confluent (yy). Isolated re- 
mains of {ee) subsequently passed into 
{see) as in Sclavonic ; but the e which 
arises from reduplication need not ne- 
cessarily be long. As regards the mu- 
tation of the short vowels, the change 
of (a) into (e), and of (o, w), into (y) is 
clear, but the mutation of (a) into (o) 
through the action of a following (u) 
or (o) is more obscure. "We can theo- 
retically assume an earliest period in 
which (a) remained pure, but it does 
not agree with the period of existing 
monuments. Hence we allow (a) to 
pass into (o) but entirely reject the 
usual assumption of the generation of 
of an — impossible — (oe) from (o) . The 
division {Brechung) of short {e) into 
(ia) and by mutation (io), must also be 

observed. As regards the consonants 
we assume h and s, here as elsewhere, 
to have been (kh, sj), though we write 
(s). The z was an abbreviation, gene- 
rally for (ts), occasionally for (st), and 
by mistake for other combinations ; the 
first alone must be retained. The \ is 
initial as in Gothic, but medially and 
finally it is softened to ^ ; as this also 
happens in most cases to the modern 
Danish d, both classes must be dis- 
tinguished from out of the corruption 
of writing. This is the weakest point 
in northern philology. The old runic 
alphabet has only the aspirate }> (th) 
and this is used medially even in the 
oldest manuscripts. The modern Ice- 
landic and Danish ^ (dh) is on the 
contrary not an aspirate but a spirant, 
which is more naturally developed from 
(d) than from (th). But since Scandi- 
navian orthography is here irremedia- 
bly confused, nothing remains but to 
restore the old essential organic ]? in 
all places where it is required by Gothic, 
Anglosaxon, and Friesic, and in other, 
partially doubtful cases, to leave d, so 
that the modern "S is altogether elimi- 
nated. The tt, which arose from an 
older (kht), must certainly be sharpen- 
ed, [that is, make the preceding vowel 
short], since reduplication can mean 
nothing but confluence ; the prolonga- 
tion of the vowel in this case is a mo- 
dern corruption, which even Grimm 
has overlooked, and similarly before ng, 
nk, and ^followed by a consonant, etc." 


When Icelanders first employed the Latin alphabet they had no 
written literature at all, and consequently no traditional ortho- 
graphy to transliterate, that is, no theoretical guide to mislead 
them. They had therefore, no means of writing except by ear, 
using the Latin letters in their accepted values, and modifying 
them for new sounds. Under such circumstances, it is scarcely 
possible that they should have — 

1 ) expressed one sound in two ways, as in the modern identities 

i ij, ^>, ei ey. 

2) made a represent (au) to the exclusion of au, 

3) have used au to express a sound (oe*) for which they had a 

form to hand, namely oi, unless indeed they had read in 
Grimm's Deutsche Grammatik (I', 474), that old Korse au 
corresponds to Gothic au, and had foreseen that the sound 
(au) would have been preserved in the German of the xix th 
A comparison of the old cognates shews that the difference be- 
tween a a, e e, etc., was originally purely quantitative. In modem 
Icelandic, as in Modern German, all short vowels before single 
consonants have become long, but in old German the length or 
shortness of a vowel was quite independent of the following con- 
sonant, as is proved by the metrical laws. In the same way the 
non-accentuation of fa^ir, father, in Icelandic originally meant 
that the vowel was short, and the accentuation of mo^tr, mother, 
that the vowel was long, as in Latin pater, mater. ^ If this view 
be well founded, the vowels in each pair, as^, a, e, e, etc., must have 
had the same quality, but different quantities, a, e, etc., being 
always short, and a, e, etc., always long; and diphthongs must 
have had the sounds of their elements connected by the glide. The 
following sounds appear then to be the only possible. 


(a, a) 

au (au, flu) 


(e, e, e) 

ei (ei, gi, ei) 


(i> »■) 

ey (ey, ei, e?) 

(O, 0, A, o) 

(B (ai, d\) 


(u, u) 

or (ee, ee^ ee) 


(y> i> ^) 

ce, (oi, ot, oe, o^, oe) 

some modification of 


or lengthened o 

The two principal criteria for selecting the correct vowel are — 

1) The palatisation of li, g, and 2) the action of vowel-mutation 
or Umlaut, (um*laut). 

^ As Icelandic still possesses really long "by position" before two conso- 
doubled consonants, the device of nants, the length of the vowel being 
donbling the consonant to indicate the confounded with the length of the 
brevity of a preceding vowel was not syllable ; but the Latins no doubt dis- 
likely to occur to the writer. That the tinguish est, is, from est, eats, as (est, 
length of a vowel depends in any way eest), and the old school joke: Mea 
upon the number of iollowing conso- mater est mala stis, could not have been 
nants is a delusion, to be classed with the ambiguous to a Latin, who would have 
notion that all vowels under the stress probably distinguished the two mean- 
must be long, and deducible probably ings as (mca maa'ter est mal"a suus ; 
from the false statement in Latin pro- me-aa maa*ter, eest maa*la suus.) — 
sodies, that a short vowel might become A. J.E. 


1) The palatisation of h, g, from (k, g) into (kj, gj) naturally 
takes place before front vowels (p. 13), while these consonants 
remain unchanged before back vowels (p. 13). Existing habits as 
to palatisation would hence determine 

e, i i, y y, m*, ei, ey to he front vowels, and 

a a, 6, u* ii, o* aw* to be back vowels, 
whereas those marked ^^ transgress this rule, <5=(aa«) commencing 
with a hack vowel, and w, o, au = {$, oe, oeoe^) with a front vowel. ^ 

2) Yowel mutation is the result of the partial assimilation of two 
vowels, not in juxtaposition, but in consecutive syllables, whereby 
the first or accented vowel becomes modified in the direction of the 
second. This may be expressed by such a formula as (a . . i = e), 
meaning that (a) in the first syllable acted on by (i) in the second 
is converted into {e). The original sounds of these mutated vowels 
or mutates, have been so changed in Icelandic, that it is necessary 
to examine the other Teutonic languages where they are better 

ia . .i=E, e), giving (e) ; old Ger. hari (nar-i), modern G. heer (heep) army. 
(*■ . . a = e, e), giving {e) ; Gothic niman (nim-an) modern G. nehmen (nee-men) to 

take ; the (i, e) forms are confused in modern German, 
(o .. i=9h, 9, i), giving (^h) ; old Ger. sconi (skoo*ni), mod. G. schon (shoea)n) 

(u .. a = o, a), giving (o) ; Gothic stulan (stuban), mod. G. ge-stohlen (ge-shtoo*- 

len), stolen, 
(u .. i = i), giving (i)) ; old G. sundia (sund'ja), mod. G. sunde (z}Tid-e) sin. 

In Icelandic we find, her^ nema, stolinn, synd (neer, nee*ma, 
stool'm, smd) all with mutates. The equation of the last word 
with modern pronunciation is (u . . i = i) which is not a mutation 
at all. The old sound must have been (i) or (y), as these are 
the only possible intermediates. The vowel mutation also proves 
that the modem sound of (B is inorganic. 

(aa .. i = EE), old Ger. wuri (bhaa'ri), Icel. vceri. 

(oo .. i = 9h), Gothic ? forjan (foor'jan), Icel. /c?r«, old /(era. 

The genuineness of the sound (oe) is made doubtful by the non- 
palatalisation of ^, and this doubt is confirmed by the equation 
(a..u=o), as in dogum for dagum. As both vowels are back, the 
result cannot be front. And the back sound of u is shewn 1) by 
the preservation of that sound in long w, 2) the nonpalatisation of 
h before it, 3) the vowel mutation. The a. . u=6, is merely a 
reversal of (u . . a=o) in stolinn, ge-stohlen, and both are quite paral- 
lel with (a.. i=e, i..a=e). 

The above conclusions result from the structure of the language, 

1 The remarks on p. 206 shew that (?, oe), as we see from the fact that 
this criterion cannot he relied on so although both sounds are used in dif- 
far as se is concerned, and, indeed, the ferent parts of Germany for o, which 
palatal action of a) on k, g, while a, a, is also frequently called (ee) or {ee), 
produced no such action, may have yet the k, g, of k'dnig, Goethe, are never 
arisen from the anticipatory action of the palatalised. This criterion can there- 
second element («). Nor is there any fore only furnish an a priori proba- 
organic necessity for the palatalisation bility. — A. J. E. 
of k, g, before such obscure vowels, as 



Chap. V. 

the following is almost positive evidence of the usages of the xn th 
century, porodclr, the grammarian, circa 1160, remarks on the 
necessity of an A, B, C, and after stating that the English have 
made an alphabet for themselves by adopting or modifying the 
Latin letters, he proposes to perform the same service for his 
countrymen — oss rslendingum^ saying : 

** To the five original Latin vowels a, e, z, o, w, I have added 
four : [now o], e [now e, ce'], (f> [now oe, as], y [now y]. Of these 
has the curve of a and the ring of o, because it is blended of their 
two sounds, being pronounced with a less open mouth than a, but 
a more open mouth than o ; ^ has the curv^e of a and the whole 
figui'e of e, for it is composed of these two, being pronounced with 
a less open mouth than a^ and with a more open mouth than e ; (f) 
is composed of e and o, being pronounced with less open mouth than 
e, and with more open mouth than o ; and y is composed of i and u, 
being pronounced with less open mouth than i and with more open 
mouth than w." 

He proceeds to give examples, shewing that e and e short cor- 
respond to modern ^, e long to modem e, e long to modern ce, o to 
modern o, g to modern o, and ^ to ce now ^. And then he remarks 
that each of these vowels begets another by beiag sounded in the 
nose, which he marks by a point above the letter. This probably 
corresponds to the palaeotypic {^), not to (a). It is now quite lost. 
Hence Rask's imaginary nasality, supra p. 550, 1. 3. 

poroddr further states that each of these 18 vowels can be long 
or short, and proposes to mark the long vowels with an accent. His 
examples shew that he places this accent in those places where an 
accent (indicating a diphthong in the case of a, 6), now exists in 
Icelandic. Then he concludes by enumerating the diphthongs, 
describing accurately the nature of diphthongs in general. Among 
these diphthongs appear au, ei, ey, but not a, d. 

The older MSS. follow poroddr with some variations. Thus the 
diacritic is often written as a full letter, as ao for o, ae for e whence 
modern ^, and the diacritic is not unfrequently entirely omitted, so 
that (9, 0, are confounded with |, g. 

The following examples shew poroddr's spelling compared with 
that now used, and the probable corresponding pronunciation. 
Abhreviations — p. poroddr's spelling, M. modern spelling, OP. old 
pronunciation, MP. modern pronunciation. 















a, a 

a aa 




oe oece 





aa, aa 

au aau 











e, ee 






9 9d 






e, ee 






u uu 












i ii 






ai aai 






i ii 






i ii 






oei cecei 




i ii 






e\ eei 







ei eei 








The sound of tlie various ^'s is evident from tlie remark that e is 
pronounced more openly than e, and more like a. The higher 
sound was given to the i as mer, German mir. The other e was an 
a, v'enia, old German wanian. In ol, the anglosaxon ealu explains 
the vowel mutation. In von the o is a mutate of a, produced by 
the preceding v, and the pronunciation has been preserved un- 
changed. The ey is a mutate of au, ^^yr«= Gothic hausjarif thus 
{au . . i=Ei) the (i) soon drawing up the (e) to {e). 

Modern Changes. — The change of (ee) to (ai) is merely the con- 
verse of the Latin <e to e} 

The a {aa) was first rounded (oo) and then broken up into (aau), 
as is shewn by the occasional MS. spelling o for a. 

The change from back (o) to front (oe) is paralleled by the English 
and most modern Danish pronunciation of (ae) for (a). 

The au changes are very complicated. Pirst, the a was rounded 
by the u into (o), as appears by the MSS. shewing gu, aou, ou for 
au. I!^ext the resulting first element, being now identical with q 
(o) was, with it, changed from back to front, into 6 (ce). Lastly 
the second element u (u) was changed by the action of the new 
front element (oe) into some front element as (i) which finally 
became (i). Thus we have the stages (au . . ou . . oeu . . ceut . . oei), 
where (oeu, oem), represents Eask's conjectural forms. 

poroddr counts U, nn, among the doubled consonants. He allows 
a double final consonant, which of course must have been a length- 
ened or 'held' consonant (supra p. 52), as in hann=(B:3imi), not 
(nan). He writes ]? everywhere, to the exclusion of ^, but whether 
this establishes a uniformity of pronunciation is very questionable. 

The following few lines will give a notion of this conjectured 
ancient pronunciation, which is placed under the present ortho- 
graphy, a verbatim translation being also interlined. 

Haustlong. (Haustlo^qg.) Autumn-lotig {night). 

E^r of-ser, er iotna otti let ofsottan 
(Edh*r ov-s^^r, by: iot'na ooi'ie \eet ovsoot'ta^n) 
Again thou-seest how of-the-giants the-terror let-sought 

Hellisbror a hyrjar haug Grjotuna bauge; 
(Hel'lesbror aa^ Hyr'iar naug Grioo'tuu^na bau'ge ;) 
Of-the-cave-the- dweller in of-fire the-hill of-Griotmi with-ring 

'Ok at isamleiki Jar^ar sunr, en dundi 
(Ook at ii'sarnbike lardh-ar su^nr ^^n du^n'd^) 
Drove to the-iron-play Earth's sun, and resounded 

Mo^r svall Meila bro^ur manavegr und hanum. 
(Moo'dhr swall M^iMa^ broo'dhur maa^n'a^w^gru^ndHa^n'u^m.) 
Bage swelled Meili's of-the-brother moon-way under him. 

1 This converse action is rare, but supra p. 294, bottom, and note 2, and 
we have a hving Enghsh example, p. 454, note 1. 



Chap. V. 

Knattu 611 en Ullar cndilag fyrir magi 
(Knaat'tu^ oil eji uMar e^n'd^laag fjT'ei>*maai'ge) 
Could all and Ullr's under-lying before the-kitisman 

Grund vas grapi hrundin ginnungave brinna; 

(Gru^nd was grap'^ rliu^nd'^jii gi^n'nuu^qga-w^e bri^n'na^ ;) 
The-ground was with-storm shaken the-wide-dwellings burn ; 

pa-es hofreginn hafrir hogrei^ar fram drogu 
(Dhaa-^s Hovreg«9^nn navr^r noog'mdhar fra^m droo'guj 
When the-temple-god the-goats of -the-elegant-chariot forwards drew 

Se^r gekk Svolnis ekkja simdr at Hrungnis fundi. 
(SedliT ge^kk Swoel'nes ckkia su^ndr at Rhim^qg-n^s fUjn*d(9.) 
Nearly went Svolnir's wife asunder to Hriingnir' s meeting [find).^ 

3. Gothic. 

In order properly to crown the edifice of the low German and 
Scandinavian dialects, it is necessaiy to consider the pronunciation 
of TJlfilas as collected from his Gothic translation of the Testament, 
etc. Grimm, Rapp, Gabelentz and Loebe, and Weingaertner,^ 
are the principal authorities. Prom a study of these works and 
the grounds on which they rely, I have arrived at certain conclu- 
sions of my own, which must be understood as referring to the pro- 
nunciation of Gothic at the time of TJlfilas, considered as a com- 
paratively modem stage of the language. There are good etymolo- 
gical grounds for believing that many Gothic words containing ai, 
aUj iu had at some previous time, a different sound from that which 
I have assigned, as for instance (ai, au, I'u), supra, p. 236, note 1. 
Eut details are here purposely omitted. The following table con- 
tains the opinions of the writers cited, as nearly as I could appre- 


1 The title means Autumn-long, 
long being the fem. of the adj. longr ; 
n6tt= night, seems to be understood; 
compare the similar old German phrase 
"den sumerlangen tac," the summer- 
long day. None of the editors trans- 
late the word, and they seem not to 
understand it. The subject of the 
poem is a fight between the god porr 
and the giant Hrungir. The poet 
describes the fight as depicted on a 
shield. The meaning of the passage, 
which is very obscure in the above 
verbatim translation following the in- 
verted order of the poet, seems to be 
as follows : Again thou seest [on the 
shield] how the terror of the giants 
[meaning porr], let sought \let ^oxi- 
phrastic=v«5?7ef/] the cave-dweller in 
the Griotun-hill with a ring of fire, 
[porr's chariot was accompanied with 
thunder and lightning] ; Earth's sun 

[that is, porr] drove to the iron-play 
[fight], anger inspired Meili's brother 
[another name for porr], and the moon- 
way [ = earth] resounded under him. 
All the wide dwellings [ =the air] could 
burn [burned], and the ground lying 
beneath was shaken with the storm 
before the kinsman of Ullr [porr again]: 
Svolnis wife [ekkja literally ividow = 
earth] nearly went to pieces, when the 
goats di-ew forward the temple-god of 
the elegant chariot to meet Hrungnir. 
2 /. Grimm, Deutsche Grammatik, 
l\ 33-74 ; P, 39-71 ; M. Rapp, Phys. 
d. Spr., i 371-401 ; Br.H.C. von Gabe- 
lentz tind Dr. J. Loebe, Grammatik der 
Gothischen Sprache, 1846, pp-22-52. 
Wilhelm Weingaertner, Die Aussprache 
des Gothischen zur Zeit TJlfilas, Leip- 
zig, 1858, pp. 68. This last work con- 
tains complete references to all the 
former essays and books on this subject. 

^ 4, No. 3. 



ciate their meaning, a (?) indicate the chief points of doubt. The 
transcription used is that employed in Gabelentz and Loebe's well- 
known edition but the letters are arranged in the order of the 
Roman Alphabet, reckoning J? as th. Leo Meyer's work (Die 
Gothische Sprache, ihre Lautgestaltung u.s.w.) came to hand too 
late to be consulted in the construction of this table. 

The Gothic Alphabet of Ulfilas. 

Abbreviations. — G Grimm, G L Von Gabelentz and Loebe, E Ellis, L letters, 
R Eapp, "VY "Weingaertner. 










G L 
















e, ee 

e, ee 

e, ee 

e, ee 























0, 00 

0, 00 

0, 00 

A, AA 



































r, r 











































H, kh 















0, uu 



u, uu 



bh, V 
ph, HV ? 

u, uu 


i, i 







i, ii 

i, ii 













In order to compare this dialect with the related Anglosaxon and 
Icelandic, I annex the conjectured pronunciation of the same para- 
ble that was selected for examples in those languages. This is also 
the same example or Gothic as Dr. Rapp has given. The verbatim 
translation is, as before, intended merely to shew the grammatical 
signification of each word. 

GotJdc, Lucas 15, 11-32. 

11. manne sums aihta tvans 

12. jah qa]? sa juhiza ize du 
attin. atta. gif mis. sei undrin- 
nai mik. dail aiginis. jah dis- 
dailida im sves sein. 

Conjectured Pronunciation. 

11. Man*n^<9 sums ekht'a 
twans suu'uns. 

12. Jakh k^^'ath sa jukh'iza 
iz*^(9 du at "tin : At'ta, gif mis, 
sii und*rin*nee mik deel eeg'inis. 
Jakh disdeel'ida im sw^^s siin. 

Verbatim Translation. 

11. Of-men certain owned two sons. 

12. Eke quoth the younger of- them 

to dad : Dad, give to-me, which unto- 
runs me, deal (part) of-ownings. Eke 
asunder-dealed to-them property his. 




Chap. V 

13. jail afar ni managans 
dagans bralita sainana allata sa 
juliiza suiius jail afiai]? iu land 
fairra visando jah jainar distah- 
ida Jjata svcs seinata libands 

14. bijje J7an fravas allamma. 
var]? hulu'us abrs and gavi jaina- 
ta. jail is dugann alafarba vair- 

15. jah gaggands gahaftida 
sik sumamma baurgjane jainis 
gaujis. jail insandida ina haij^jos 
seinaizos lialdan sveina. 

16. jah gaimida sad itan 
haurne. foei matidedun sveina. 
jah manna imnia ni gaf. 

17. qimands ]7an in sis qa]?. 
wan filu asnje attins meinis 
nfarassau haband lilaibe. i]? "ik 
huhrau fraqistna. 

18. usstandands gagga du 
attin meinamma jah qi]7a du 
imma. atta. fravaurhta mis 'in 
himin jah in andvair]7Ja feinam- 

19. ju ]7anasei]7s ni im vairj^s 
ei haitaidau sunus Jeins. gatavei 
mik sve ainana asnje J^einaize. 

13. Jakh af'ar ni man'agans 
dag'ans brakht'a sam'ana aMata 
sa juklriza smrus, jakh afleeth' 
in land fer'ra wis'andoo, jakh 
jcen'ar distakh'ida that*a sw^^s 
siin'ata lib'ands usstyyr'iba. 

14. Bithee* than frawas" al*- 
lam'ma, warth H'uukh'rus ab''rs 
and ga'wi jeen'ata. Jakh is 
dugan* al'atharb'a werth'an. 

15. Jakh gaq'gands gan'aft*- 
ida sik sum'am'ma bArg'janee 
jeen'is gAA'jis. Jakh insand'ida 
in'a H'eethjoos siin'eezoos H'ald'- 
an swiin'a 

16. Jakh gem'ida sad it'an 
B:Ar'jiee, thoo'ii mat'id^^d'un 
swiin'a. Jakh man'na im'ma ni 

17. k^^Jim•ands than in sis, 
ki^ath : K«<;han fil'u as'iuee at'- 
tins miin'is uf'aras'SAA Hab'and 
khleeb'^^, ith ik H'uukhTAA 

18. Us* stand *ands gaq'ga du 
at'tin miin'am'ma, jakh ke^ith'a 
du im'ma : At'ta, frawArkh'ta 
mis in H'im'in, jakh in and'- 
werth'ja thiin'am'ma, 

19. JU than'asiiths ni im 
werths ii n'eet'eedAA sun'us 
thiins ; gata'wii mik swee een*- 
ana as'nj^i? thiin'eez^^. 

Verbatim Translation. 

13. Eke after not many days 
brought together all the younger son, 
and off-led (departed) in land far being, 
eke yon asunder-tugged (dissipated) 
the possession his, living out-steeringly. 

14. By-that then from- was of- all, 
worth (became) hunger strong against 
region yon, eke he began quite-needy 
to-worth (to-become). 

15. Eke ganging joined himself to- 
certain of-burghers of-yon region ; eke 
in-sent him of-heath his to-hold swine. 

16. Eke yearned fall to-eat of-horns 

(husks), which meated (ate as meat or 
food) swine ; eke man to-him not gave. 

17. Coming then in himself, quoth: 
How many hirelings of-dad mine in- 
overmuch (abundantly) have of-loaves, 
but I by-hunger perish. 

18. Out-standing I -go to dad mine, 
eke say to him, Dad, I-from-wrought 
(I -sinned) for-me in (against) heaven 
eke in face thine. 

19. Now the-since (longer) not am 
worthy that I-may-hight son thine ; 
y-do (make) me as one of-hirelings 

§ 4, No. 3. 



20. jah usstandands qam at 
attin seinamma. nauh]7arLuh j^an 
fairra visandan gasaw ina atta is 
jah infeinoda jah jjragjands draus 
ana hals is jah kukida imina. 

21. jah qaj? imma sa sunus. 
atta. fravaurhta "in himin jah in 
andvaii^fja Jjeinamma. ju f ana- 
sell's ni im vairjjs ei haitaidau 
sunus j^eins. 

22. qa]? fan sa atta du skalkam 
seinaim. sprauto briggi]? vastja 
]70 frumiston jah gavasji]? 'ina 
jah gibi]? iiggragul]? in handu is 
jah gaskoh ana fotuns is. 

23. jah briggandans stiur 
jjana alidan ufsnei]?!]?. jah mat- 
jandans visam vaila. 

24. unte sa sunus meins dauj^s 
vas jah gaqiunoda jah fralusans 
vas jah bigitans varf . jah dugun- 
nun visan. 

25. vasujjfan sunus is sa 
alfiza ana akra jah qimands at- 
iddja new razn jah gahausida 
saggvins jah laikans. 

26. jah athaitands sumana 
magive frahuh. wa vesi ]?ata. 

20. Jakh US" stand 'ands kt^am 
at at'tin siin'am'ma. NAkh'- 
than'ukh than fer'ra wis'andan 
gasakt^^h" in*a at'ta is, jakh 
infiin'ooda jakh thrag'jands drAAs 
an"a n'als is jakh kuk'ida im*ma. 

21. Jakh kw;ath im'ma sa 
sun'us : At'ta, frawArkht'a in 
Him "in jakh in and'werth'ja 
thiin"am*ma, ju than'asiiths ni 
im werths ii H*eet*eedAA sun'us 

22. 'Kwsith. than sa at'ta du 
skalk'am siin'eem : SprAAt'oo 
briq-gith wast'ja thoo frum*- 
istoon jakh gawas'jith in*a jakh 
gib'ith fiq'gragulth in s'and'u is, 
jakh gaskookh* an'a foot'uns is. 

23. Jakh briq'gandans styyr 
than-a al'idan ufsniith ith, jakh 
mat'jandans wis-am weel'a. 

24. TJn-t^^, sa sun'us miins 
dAAths was jakh ggikwyyn'ooda, 
jakh fralus-ans was jakh bigit*- 
ans warth. Jakh dugun*nun 

25. Was'uth'than sun-us is sa 
al'thiza an*a ak'ra, Jakh hwim.'- 
ands at'iddja nekwh. raz-n, jakh 
gan'AAS'ida saq'gwins jakh 

26. Jakh at's'eet'ands sum*- 
ana mag-iw^^ frakhukh, k«<;ha 
w^es'i that'a. 

Verbatim Translation. 

20. Eke out-standing came to dad 
his ; still then far being saw him dad 
of-him, eke pitied, eke running fell on 
neck of-him, eke kissed him. 

21. Eke quoth to-him the son, Dad, 
I-from-wrought (I -sinned) in (against) 
heaven eke in face thine. Now the- 
since (longer) not am worthy that 
I-may-hight son thine. 

22. Quoth then the dad to servants 
his, Quickly bring vest the from-est 
(first, best), eke in- vest him, eke give 
finger-gold in hand of-him, eke shoes 
on feet of-him, 

23. And bringing steer the fatted 
up-cut, eke meating (eating food) let- 
us-be well. 

24. Unto-that (because) the son 
mine dead was, eke y- quickened, eke 
lost was, eke be-gotten worth (became). 
Eke they-began to-be (to feast). 

25. Was-then son of-him the elder 
on acre, eke coming to-went (ap- 
proached) near house, eke heard song 
eke games. 

2G. Eke to-calling certain of-boys, 
asked, what were that. 



Chap. V. § 4, No. 3. 

27. ]?aruli is qaj? du imma. 
fatei brojjar jjeins qam. jah 
ufsnaij? atta feins stiur fana 
alidan. unte hailana ina and- 

28. fanuh modags varj? jah ni 
vilda inngaggan. 'i]> atta is us- 
gaggands ut bad ina. 

29. fanih 'is andhafjands qaj? 
du attin. sai. sva filu jere skalk- 
inoda ]7us jah ni wanhun an- 
abusn J?eina ufariddja. jah mis 
ni aiv atgaft gaitein ei mij? fri- 
jondam meinaim bivesjau. 

30. "if fan sa sunus feins. 
saei fret J^ein sves mi]? kalkjom. 
qam. ufsnaist 'imma stiur jjana 

31. ]7aruh qa]? du imma. bam- 
ilo. |7u sinteino mi]? mis vast jah 
is. jah all fata mein fein ist. 

32. vaila visan jah faginon 
skuld vas. unte brofar feins 
daufs vas jah gaqiunoda. jah 
fralusans jah bigitans var]?. 

27. Thar'ukh is kw^ath du 
im'ma : That-ii brooth-ar thiins 
kM;am, jakh uf'sneeth* at*ta 
thiins styyr than*a al'idan, un't«(9 
H'eel"ana in*a andnam*. 

28. Than-ukh mood'ags warth, 
jakh ni wil'da in-gaq-gan. Ith 
at'ta is us'gaq'gands ut bad in'a. 

29. Thar-ukh is andn'af-jands 
kt^^ath du at 'tin : See, swa fil*u 
jeevee skalk"inood*a thus, Jakh 
ni ki^han'H^un an-abus'n thiin'a 
uf'ar,id*dja. Jakh mis ni eew 
at'gaft. geet'iin ii mith fri*- 
joond-am miin-eem biw^^s'jAA. 

30. Ith than sa sun'us thiins, 
sa,ii' fr^^t thiin sw^^s mith 
kalk'joom, kwsiWi, uf'sneest' im-- 
ma styyr than'a al-idan. 

31. Thar'ukh kz^^ath du im*- 
ma : Bam'iloo ! thu sint'iinoo 
mith mis wast jakh is ; jakh al 
that'a miin thiin ist. 

32. "Weel'a wis -an jakh fag*- 
inoon skuld was un*t^^ brooth'ar 
thiins dAAths was Jakh ga- 
k«^;yyn'ooda, jakh fralus'anz 
jakh bigit'ans warth. 

Verbatim Translation. 

27. Then he quoth to him, that 
brother thine came, eke up-cut dad 
thine steer the fatted, unto-that (be- 
cause) whole him received. 

28. Then moody worth (became), 
eke not would in-go. But dad of-him 
out-going out had him. 

29. Then he to-heaving (answering) 
quoth to dad, Lo, so many years served 
to-thee, eke not whenever command 
thine over-went (transgressed), eke to- 
me not ever at-gave goat, that with 
friends mine might-feast. 

30. But then (when) the son thine, 
who devoured thine possession with 
harlots, came, thou-up-cuttest for him 
steer the fatted. 

31. Then quoth to him, Little-son, 
thou always with me wast, eke art, eke 
all the mine thine is. 

32. "Well to-be eke to rejoice due 
was, unto-that (because) brother thine 
dead was, eke y- quickened ; eke lost, 
eke be-gotten worth (became). 

Chap. VI. § 1. VALUE OF LETTERS. 565 


On the Correspondence of Orthography with Pro- 
nunciation FROM the AnGLOSAXON TiMES TO THE 

Present day. 

§ 1. The Value of the Letters. 

The several conclusions arrived at respecting the sounds 
of the letters in English orthography are necessarily very 
irregularly scattered through the preceding pages. The 
nature of the investigation which obliged us to commence 
with the XVI th century, then descend through the xvii th 
to the xviii th, and immediately jump to the xiv th, and then 
after a glance at the xv th, commence the consideration of 
the XIII th century, has not produced an order which is con- 
venient or satisfactory to the reader. In the present section 
then the results will be arranged in a tabular form, in alpha- 
betical order. A reference to the pages in which the several 
statements are established, is occasionally given, but as it 
was found impracticable to introduce it concisely into the 
text in all cases, the indices at the end of the book must be 
consulted. The outline index annexed will enable the reader 
to refer immediately to the principal combinations. 

The construction of the Table is as follows. All the single letters 
or combinations of letters which have been used as parts of words 
in English orthography, from the Anglosaxon period to the present 
day, such as a, aa, cb, ae, a-e (meaning a followed by some consonant 
and then by e final), of, -age (meaning age final) ah, ai, al, all, an, 
-ange, ao, aou, ar, as, -aste, ath, au, augh, aun, aw, aww, ay, ago, I, 
etc., are placed in alphabetical order at the head of separate para- 
graphs, as in a dictionary, and then the history of the different 
sounds that each has represented is sketched in accordance with 
previous results, using 

ags., for the Anglosaxon period, 

13., for the xrcith century and earlier. 

14., 15., 16., 17., 18., 19., for the xivth, xvth, xvith, xviith, 

xvm th, and xix th centuries respectively. 
The passages inserted in brackets at the end of some articles, 

signed P., are due to Mr. Payne, see infra, pp. 579-80. 



Outline Index to the Principal Combinations. 
Anglosaxon period \ p. 510. 

Thirteenth Century and Earlier: pp. 423, 431, 439, 467, 471, 476, 480, 484, 
487, 496, 498, 506. 





14. 259, 

14. 263, 



14. 316, 

16. 59, 

16. 118, 


17. 18. 188. 

16. 17. 18. 196. 

17. 65, 

17. 124, 



18. 74. 

18. 129. 



14. 317, 




17. 18. 188. 

16. 17. 18. 214. 

14. 263, 

14. 260. 



16. 118, 




14. 317, 

17. 124, 

14. 301, 


17. 18. 188. 

16. 17. 18. 214. 

18. 129. 

16. 136, 137, 



17. 139, 

18. 141. 

14 808 



14. 317, 

14. 263, 



16. 17. 18. 203. 

16. 136, 141, 




17. 147, 

16. 17. 18. 219. 



14. 317, 

18. 149. 


16. 17. 18. 219. 




14. 308, 




14. 308, 

16. 17. 18. 203. 



14. 298, 

16. 17. 18. 203. 


X f • 



16. 160, 163. 


14. 308, 

14. 310, 

16. 17. 18. 209. 




17. 171, 

18. 184. 

16. 17. 18. 203, 


01 OY 


14. 308, 
16. 17. 18. 203. 

14. 308. 
14. 314, 
16. 17. 18. 220. 

I, Y 
14. 270, 

16. 104, 

17. 116, 




14. 269, 

16. 17. 18. 135. 


14. 317, 



16. 17. 18. 219. 

14. 308, 

16. 17. 18. 203. 




14. 317, 


18. 117. 



16. 17. 18. 184. 

14. 260, 318, 





16. 77, 

14. 260, 


14. 317, 

17. 81, 

16. 104, 



16. 17. 18. 184. 

18. 88. 

17. 116, 


136, 149, 



18. 117. 



14. 317, 

14. 260, 




16. 17. 18. 214. 

16. 77, 

17. 81, 

14. 314, 

16. 17. 18. 203. 




Y vowel, see I 

18. 88. 



17. 18. 203. 

Y consonant. 


14. 315, 


14. 310, 317, 

14. 260, 

16. 17. 18. 203. 



16. 17. 18. 184. 

16. 77, 




17. 81, 

14. 315, 



14. 310, 317, 

18. 88. 

16. 17. 18. 193. 


17. 18. 203. 

16. 17. 18. 214. 

Chap. VI. § 1. 




Chronological Account of the Values of Letters. 

A ags. was both a short and a long 
sound \a, aa), but the long sound was 
sometimes written d. Short a in an 
open unaccented syllable was probably 
(a). After ags,, a in an open accented 
syllable was considered as long, and in a 
closed syllable generally short. In 13. 
14. 15. 16. a seems to have been (a, aa), 
although in the earlier part of this time 
it may have been {a, aa). Probably 
towards the end of 16. it passed into 
(ah, aah), a sound frequent in 19. In 
17. it became (se, sese), and at the latter 
end of 1 7. and beginning of 1 8., it seems 
to have fallen into (se, ee). These 
changes seem to have occurred towards 
the close of 15. or even earlier in Scot- 
land, p. 410, n. 3, and perhaps in the 
Northern and West Midland Counties, 
p. 450, n. 2. See references under 
ey. Perhaps during the latter part 
of 18. there came into use a distinc- 
tion, thoroughly established in 19., 
that long a should be {ee) unless fol- 
lowed by r, and that then it 
should be (ee) ; compare naming, Mary 
(mem*eq, MeeJ[•r^). In 19. long a is 
frequently pronounced {ee\) in place of 
{^ee\ as (nmm) for (ne^m), pp. 234, 272, 
n. 3 ; 294, n. 2. Short a has re- 
mained (se) from 17. to 19. These 
general usages have been crossed by 
the action of a following /, w, r, s, /A, 
see af, an^ ar, as, ath, and the other 
combinations which follow. An initial 
(w) acted in the latter part of 17. and 
subsequently, in many, but by no means 
all words, to convert (se) into (a) or 
(a), as in was, what, etc. In 19. a has 
been variously degraded as in : hating, 
father, water, many, hat, want, riband 
= {neet'iq, faadhu, WAAtu, men-e, Hset, 
WAnt, rib "en). 

A A was in 14. occasionally used for 
(aa) in closed syllables. Otherwise it 
was only employed in biblical names, 
as Aaron, Isaac, and then it followed 
the sound of long or short a. It was 
occasionally for German aa, and then 
from 17. it was (aa). 

M ags. (se, sese), p. 510, in 13. sank 
to (e, ee) or (e, ee), pp. 487, 496, 
498. It was rarely used in 13., and 
not at all afterwards, except in words 
borrowed from the Latin or Greek, and 
then it was (ee) till 18., towards the 
close of which it became (ii) in such 
cases. But scholars still occasionally 
say (e) as in : Psestum = (Pest'Bm) 

rather than (Piis-tem), which is also 
heard. [In 13. <e = (ee) in Norman 
and English.— P.] 

AE was never an English combina- 
tion, but, resulting from biblical names 
or Latin adaptations, it seems to have 
been treated as a+e, or se. In 19. we 
have aerie, Michael, aerial, Israel = 
{eex-ri iirri, Mgi-k^l, e,evivl eei-ri-el 
iii'ri^l, Iz'reel Jz'rel /ztbI). [In 13. 
ae=se = {ee) in Norman and English. 

A — E, that is a followed by some 
consonant and a final e, which, when 
pronounced, had the effect of putting a 
into an open syllable, and therefore 
making it long, so that when the final e 
ceased to be pronounced, it was presumed 
to have the same effect of lengthen- 
ing the preceding vowel. Hence a — e 
was assumed to be long a, with the 
sound of the time, from 16. to 19. Per- 
haps this feeling came in towards the 
close of 15. The rule is not con- 
sistently carried out in 19, ; compare : 
hate, are, landscape, furnace, have = 
(neet, aai, Isend'sk/p, h-ays, asav). 
Even in 16. the vowel was not long in 
unaccented syllables. 

AF, this combination presents no- 
thing peculiar till 18. or 19. and then 
only in certain words : graflf, staff, dis- 
taff, quafi', aft, after, abaft, haft, shaft, 
raft, craft, draft, graft, waft, and laugh, 
calf, half, which must be considered to 
have the same combination. Here 
usage differs. The common southern 
pronunciation is (aaf), and even (aaf) 
may be heard ; the fine educated nor- 
thern pronunciation is (sef). Ladies in 
the South and many educated gentle- 
men say (ahf) or at most (aahf). But 
(af) is also heard. Those who use the 
finer sounds, ridicule the others as 
vulgar, and write them larf, etc., de- 
claring that an r is introduced, but 
this arises from their own omission of 
(r) and preservation of (aa), in : barm, 
starve, etc. See ar, or, and the cita- 
tion under o, p. 575, col. 1. 

AG in late ags. and 12. or 13. was 
probably equivalent to (ai). 

Ajt;, Orrmin's form of (ai), p. 488. 

-AGE. In 16. the ^e = (dzh) seems to 
have influenced the preceding a by in- 
troducing an (i) sound, as (aidzh), p. 
120 ; and in 17. to 19. this a has fol- 
lowed the fortunes of ai, which see. 



AH, as an exclamation, has probably 
always represented (aa), although the 
corresponding exclamation was not 
always represented by ah. In dahlia 
it is now pronounced {ee). 

AI in 14. =(ai, aai), which sounds 
apparently remained to the end of 16., 
though the pronunciation (ee) was in 
use by a large number of speakers. In 
17. after a passage through (ahi, a;i), 
the sound rapidly sank to (ee), but 
whether the sound (eei) was not occa- 
sionally heard cannot be ascertained 
with certainty. In 19., {ee, eei) are 
both usual forms. Various degrada- 
tions are heard in 19., as : dem«m, satl, 
S«mt John,s«i'd,[plfl/d, Britam (dimiin*, 
s,ee\ Sm-dzh^n, sed, plsed, Br^■t•'n), and 
dais, which was a monosyllable in 
Chaucer, 372 = (dais), but has be- 
come dissyllabic = (d6t'-is). For 13. see 
pp. 431, 440, 467. 473, 506; 14. 459, 
462 ; 15. 447. See especially p. 459, 
n. 1, and the passages there referred to, 
and also Chap. VII. § 1. The use of 
(ai) for (ee) seems fixed in Scotland at 
the beginning of 16., p. 410, n. 3. [In 
13. and 14. ai = ai/ = {ee) in Norman 
and English ; in 16. often, if not gene- 
rally = (ai) in English, infra p. 582. 

AL, ALL in 16. and hence probably 
for some time previously the I had be- 
gun seriously to influence the preceding 
vowel, by being pronounced ('1) with a 
very appreciable length of murmur or 
being labialised into Q.w) ; the result 
in either case, accepted as (ul), pro- 
duced the dipnthong (aul), which was 
firmly established in 16. See /, p. 193. 
This was occasionally followed by the 
total disappearance of the I, as in : talk, 
calm = (tank, kaum). Then this al 
was considered as tantamount to au, 
and followed its changes, becoming 
(aa) in 17- and in most words so re- 
maining to 19. ; but in some words, as 
(p«/m, calm), although occasionally 
called (pAAm, kAAM) in 17., and in 
Irish-English, p. 76, the combination 
seems to have generally resisted the 
change -"o (aa), and rather to have 
passed from (aau, aa') to simple (aa), as 
we still hear (paam, kaam), refined by 
some to (paahm kaahm, psesem ksesem, 
paem kaera) ; while others, inorgani- 
cally and purely orthographically, at- 
tempt to say (paelm, kselm). See au, 

AN. In 16. French words now 

having the nasal vowel (aA) were 
heard as having (aun), p. 143, and 
hence the writing aun much prevailed 
then ; and as we also find this ortho- 
graphy in 14., probably the same effect 
was produced on English ears by that 
French sound. In 16. aun was occa- 
sionally replaced by an, as commaund, 
command, but probably the sound (aun) 
remained. In 17. the sound became 
(aau), and during 18. and even into 19. 
this sound remains, although there is, 
and perhaps always was, a tendency to 
fall, on the one hand into (aan), on the 
other into (an), with their various re- 
finements ; see af. Thus romance 
romantic have now generally (sen), but 
(aau) is occasionally heard, and forty 
years ago I was familiar with (romAAns*, 
romaans'). In command, demand, etc., 
the contest is among (an aan, an aan, 
sen seaen, ahn aahn). In daunt, gaunt, 
haunt, gauntlet, jaunt, taunt, vaunt, all 
the last named sounds may be heard, 
and also (aau), but never (An). It 
would be convenient to use (aan) for 
{an) in all words where it corresponds 
to the modem French (aA). See au. 

-ANGE. In 16. the sound (i) was in- 
serted as (aindzh),p. 120, and the com- 
bination was treated in 17. as if written 
-ainge, the a becoming (ee) and then 
{ee) or {eei) in 19. In unaccented 
syllables it drops into (-^ndzh, or 
-mdzh) properly (-yndzh), as oranges 
= (or'yndzh^z). 

AO. This is never recognized as a 
true English combination, though it 
occurs in gaol now (dzh^d), and by 
accidental attraction in extraordinary, 
now (ekstrAAi*dmer»), and foreign 
words, as : Pharaoh, aorta, Chaos, now 
(Feei'ro, e,'di'i'Q, Ke^-os). The old pro- 
nunciation of gaol is doubtful. Extra- 
ordinary was probably always treated as 
a compound, compare " afford no extra- 
ordinary gaze," Henry IV. part 1, act 
3, sc. 2, V. 78. 

AOU. This French mode of writing 
(au) is only met with in caowtchouc, 
generally called (kautsh-wk), but occa- 
sionally (kuut'tshwk) in 19. 

AE. The vocal character of r as 
('r) seems to have acted upon the pre- 
ceding vowels in all cases after 16. 
Probably ar, when not followed by a 
vowel, remained (ar) or (ai), though 
unacknowledged, during 17. 18._ 19., 
with the variation (aai), which is in 19. 

Chap. VI. § 1. 



frequently reduced to simple (aa) . But 
ar was frequently called (sGr) or (sej) in 
17. and 18., and the sound is still heard 
in American English. In the present 
usage of the South of England the 
{x) is practically dropped, pp. 196, 245. 
See 0, or, r. 

AS. In a few words of 19. the s 
seems to react on the a, as : pass, class, 
mast, fast, in which a receives all the 
variety of sound noticed in a/, an, as 
(paes pseses, pas pass, pas paots, pahs 
paahs). In other words, as : passage 
classify, [classics sometimes follows 
the rule of class), gas, (mastiff' is doubt- 
ful), no such action takes place. It is not 
noticed by older writers, and is there- 
fore probably modern, but it may be 
merely a remnant of the 16. and earlier 

-ASTE, in 16. and earlier (ast), but 
in 19. we have : haste, paste, taste, 
waste (now distinguished from waist, 
which was not the case in 16., see p. 
73, note l)={iieest, petst, teest, weest). 
Here the action of 5 is precisely con- 
trary to that in as. No clue to this 
change has been discovered, but we may 
conjecture an intermediate (nffiaost, 
paesest) during 17. Could there have 
been an inserted i, as indicated by the 
spelling waist in one sense of 16, waste, 
analogous to that in -an(/e, ash, lash, 
pp. 120, 264 ? 

ATH. In : path, bath, lath, wrath, 
th seems to have acted as /, s (see af, 
as) in preserving the (a) sound, or its 
modern variants [a se ah), short and 
long, in 19. 

ATI. See awiv. At a very early 
period in 13. and 14. an, aw were 
(au), which sound remained to 16. 
Either at the close of 16. or beginning 
of 17. it seems to have passed through 
(rtu, aau, aa') into (aa), in which form 
it was firmly established in 17. and has 
remained with little or no change, but 
is occasionally (aa). See aun, an. In 
19. we have isolated degradations, 
compare : gauging, aunt, haul, hauteuv, 
Jervawlx, laurel, meersch«wm, Mene- 
lau& = {geedzh.'iq, aant, haaI, uootix', 
Dzhaaj'Vis, lAr-el, miij-sh^m, Meni- 
lee'ds), where the foreign words have 
received an English pronunciation. [In 
13. and 14. au generally (au), but be- 
fore n, especially in 14. = (aaa) in Nor- 
man and English, infra p. 583. — P.] 

ATJGH. This must be considered 

as a double combination au+gh, the 
first part follows au, the second gh, 
hence in 14. laugh = (laukh, lauktt^h, 
lauwh), in 16. = (laukh, lauH'), in 17. 
(laegef) or (Isef), perhaps also (laaf) as 
in 19. See af. The gh becoming 
occasionally mute, augh was treated 
altogether like au, as in : taught, 
caught = (tAAt, kAAt). 

AUN. See an. 

AW. This was precisely equivalent 
to au. In 14. it was used in the 
middle as well as at the end of a word. 
In 16. and afterwards it was seldom 
used except when final, though we still 
write : awl, awning, brawl, crawl, 
prawn, sprawl, etc. 

AWW. Orrmin's form of (au), p* 

AY. Precisely equivalent to ai. In 
14. used in the middle as well as end of 
words ; in 16. and afterwards generally 
final. See references under ai, ei. 

AYO. In the word mayox— (mee.i) 
in 19., ago may be considered as a single 
combination, but it is properly ay-\-o ; 
Mayo is generally called {Meco). 

B. Ags. to 19. = (6), but in 19. not 
unfrequently written Avhen not pro- 
nounced as in deZ>t, dou(^t, lam5, h&Ql. 
Hum, suZ»tile ; in deit, doui^t it was not 
pronounced and generally not wi'itten 
in 16., p. 211, n. 2. It was mute in 17. 
in all the cases in which it remains so 
in 19. 

BB. Like other doubled letters, had 
the sound of the single letter (b), being 
only used to indicate a preceding ac- 
cented short vowel. 

C. In ags. always (k) or {k), but at 
a later period of ags. the {k) seems to 
have become (tsh), p. 511. See ch. 
In 13. it is apparently not used before 
(e, i), except in the combination -sce= 
-sse, and then it was (s) ; but in 14. 
when French words were freely intro- 
duced it was (s) before e, i and (k) 
otherwise, and so it has remained ; but 
see ce-, ci. 

CC. In ags. the same as c, but indi- 
cating that the preceding vowel was 
short and generally accented ; in later 
times either (k) or (ks) as in : account, 
accident = (aekaunt-, sek's/dBnt) in 19. 

CCH in 14. used for fcA = (t+tsh), 
and pronounced (tsh), shortening the 
preceding vowel. 


VALUE OF LETTERS. CE — EA. Chap. VI. § 1. 

CE. Till 18. this seems to have been 
simply c + e. At the end of 17. it 
changed to (sh) in in ocean. See ci, 
si, ti. 

CH. Not used in ags., but in 13. 
found in the signification of (tsh), the 
sound into which [k) had fallen, and as 
such it has remained. In words from 
the Greek as art^itect it is (k) in 19., 
and probably was so in 14. ; in words 
from the modern French as cAaise it is 
(sh) in 19., but for French words intro- 
duced before 18. as c/min, the sound 
(tsh) seems to have prevailed. In a 
few final syllables as : Greenwich, 
Woolwich, Norwich, it has become 
(dzh) in 19., but in others it remains 
(tsh), as Ipswich, locally (/ps-/dzh), p. 
512, n. 2. In fucAsia = (fiu-shia) it is 
mute. See si-. In 13. it was rarely 
used as^/i = (kh), p. 441. In modern 
Scotch it has the three sounds (A;h, kh, 
kw^-h) determined generally by the pre- 
ceding vowel. 

CI-. Till 18. this appears to have 
been simply s-\-i^ but then it fell into 
(sh), as spec/al, spemus, offic«al = 
(spesh'^l, spii'shes, ofVsh'^l). See si-, ti-. 

CK. This means kk or {k) from 14. 
to 19., but in 14. kk is fi-equently used. 

GW in ags., p. 514, probably = (kw;) 
that is nearly (kw) ; replaced by qu 
after ags. 

CZ. This is a modern combination 
used chiefly in Sclavonic words, as 
Czech, Bohemian (tshekh), but English 
(tshek) : Czar is called (zaai) in 19., 
but its Eussian initial is (ts). 

D ags. to 19. = (d). When, how- 
ever, the past participle ed dropped its 
e, the d changed to (t) after mutes or 
hisses, as : capped, sacked, quaffed, 
kissed, at least in 17. and probably 
even in 13. as bliscedd= (blist), p. 444, 
note 2. In 19. d is palatised into (dj, 
dj), and ultimately (dzh), in many cases, 
acknowledged or repudiated, as : soldier 
= (sool-dzha),verdure = (v.i-diuj, vi-djur, 
vj-djur, vidzh-j), the last having the 
same sound as verger. It is generally 
mute in : riband, Wednesday. 

DD. Whenever used =(d), except 
in compounds. 

DG = (dzh) from 14. to 19., before 
a palatal vowel, as e, i as : ]Vidge, bri^^- 
ing and sometimes this sound is re- 
tained, even when an e has been ortho- 
graphically omitted, as ju(?^ment. 

D In ags. ^ was either (th) or (dh) 
perhaps used indifferently in the MSS. 
which we have, p. 515. In some more 
recent ags. and in 13. ^ was used as 
the only sign for both (th, dh), in 
others ]> was the only sign. After 13. 
"S seems to have been discontinued, and 
only ]) used in 14. and part of 15. 
Even in 13. th was occasionally used 
for either 'S or b. Judging by modern 
Icelandic habits "S was (dh) when 
medial or final in ags. See also p. 541, 
n. 2, p. 555, n. 1. 

E=ags. (e, ee), and this sound it 
seems to have retained to the middle of 

15. Then some of the words with e 
long had the sound of (ii), but e short 
has remained (e) to 19. The use of 
long e as (ee, ii) fluctuated much during 

16. and 17., but in 18. the sound (ii) 
established itself and has remained. 
See ea, ee. In 19. it has a few anoma- 
lies, compare : b^, clerk, prdty, lei, resin, 
hideous, opm = (bii, klaaak, pr/te, let, 
roz-in, H?d*J8s, oop-'n). Finale seems 
to have been pronounced, at least in 
the Southern parts of England, till the 
beginning of the 15. with certain ex- 
ceptions, pp. 318, 364. During 15. 
most final e's lost their sounds, and in 
16. e final was considered to indicate 
that the preceding vowel had its long 
sound. The final e seems to have be- 
come silent even in 14. or 13. in the 
northern parts of the country, p. 410. 
Usages differ in existing MSS. 

EA. In ags. this seems to have 
been a true diphthong (ea) with the 
stress generally on the first but occa- 
sionally on the second syllable, indi- 
cated by (ea, ea), p. 511. Although 
found in 13. pp. 467, 498, we may con- 
sider that with ags, it passed out of use. 
It is occasionally found in 14. as (ee). 
It was not till the middle of 16. that it 
was extensively used to mark those 
long e's which retained the sound of 
(ee) in contradistinction to those which 
had fallen into (li), the latter being 
written ee. This distinction was how- 
ever not consistently carried out even 
at first, some words having the (ii) 
sound being spelled with ea, and all 
sounds having the (ee) sound not being 
spelled with ea. In 17. still more of 
the words with ea became sounded as 
(ii) without any change of spelling, 
and by the middle of 18. the use of ea 
generally as (ii), and rarely as (ee, ee) 
as in : bear, great, was established and 

Chap. VI. § 1. 



has remained to 19. Many words in 
ea which had long (ee) in 14. were 
pronounced with short (e) at an early 
period, as : ht^fld, lead s. In the earlier 
part of 18. the sound of (ii) was applied 
to words such as great, break, which are 
now generally pronounced with {ee). 
The 19. varieties are seen in : heal, 
great, heart, guinea, head, react, area, 
= (niil, greet, naajt, gm"e, ned, rieekt", 
eei"ri,'B). [In 13. and 14. ea—ae = ai = 
(ee) in Norman and English, infra p. 
682.— P.] 

EAU. This form was not employed 
in 14., but ew was used in place of it ; 
even Levins, 1570, has bewtye. In the 
earlier part of 17. eau was (eu), in the 
later [part and since, (iu). As usual, 
19. furnishes varieties, as in : Beaw- 
champ, beaw, heaufm, beawty = (Biitsh'- 
■Bm, boo, bef*m, biu'tt). [In 14. eau 
— eal, iau in Norman of 13. =ew, etf', 
= (uu) in Norman and English, infra 
p. 586.— P.] 

EE. Invariably represented (ee) in 
14. and was generally used in closed 
syllables. At beginning of 16. it was 
sometimes (ii) and sometimes (ee). 
During the latter half of 16. it was 
fixed as (ii), the (ee) sound being gene- 
rally written ea (which see). So it 
has remained. In 19. breeches is 

E'E. A 17. and later contraction 
for eve in e'er ne'er and pronounced 
(ee) up to 19. 

E-E. The affixed mute e rendered 
the preceding e long, and hence in 16. 
the sound was generally (ee), but in 
some cases (ii) . The spelling was then 
discontinued, ea, ee taking its place, 
thus Salesbury's chepe^ chese became 
cheap, cheese. At the beginning of 18. 
the sound of (ii) prevailed and has con- 
tinued ; but 19. shews : these, there, 
allege = (dhiiz, dheej, aledzh-). 

EG- in later ags. and in 12. (ei, ai). 
E^^. Orrmin's form for (ei), p. 489. 
EH, the exclamation (ee, ee). 

EI. In 13. seems to have been (ei, 
ai). In 14. when used, which was rarely, 
ey being the common form, it was (ai) 
sometimes (aa,i) pp. 264, 476. See the 
references given under ai. In 16. it 
varied as (ei, ai), and in 17. became 
(ai) or more usually (eei, ee). During 
the latter part of 18. it changed to (ii), 
where it generally remains, with va- 

rieties of (ai, ee) as in : conceit, veel, 
forfeit, heifer, dei'pnosophist = (konsiit', 
veel, fojfit, Hef'jc, doipnos-of/st). In the 
words either, neither, ei was generally 
(ee) in 18. ; in 19. usage fluctuates be- 
tween (ii, ai), some still use (ee), 
p. 129, n. 1. [Precisely the same as 
ai, ay, infra p. 582. — P.] 

EO. In ags. this seems to have been 
generally (eo) but occasionally (e6). 
In 13. eo interchanged with e and the 
sound was (ee), p. 487. The combina- 
tion then went out of use, although 
both eo and oe are found in 14. in the 
sense of (ee). In 17. therefore it be- 
came (ii) in people, and even in yeoman, 
though this has now {oo). As eo is 
rare and has come from many sources 
it is very variously pronounced in 19., 
as : people, Georgies, yeoman, galleon, 
Theobald, leopard, dungeon, Macleod, 
ieod, theologian, theology = (pii-p'l, 
Dzhoi-dzh/ks, joo-m^n, gseluun*, Tib*- 
-eld,, dan-dzh^n, ra^klaud-, find, 
thii,oloo*dzhit;n, thiol'odzhi). [In 13. 
and 14. eo, oe= (ee) generally, but often 
= (uu) in Norman, and sometimes in 
English, infra p. 586.— P.] 

EOU, EOW, perhaps (eou) or (eu), 
p. 498. [In 13. and 14. eow in Eng- 
lish =(uu), infra p. 586.— P.] 

EE, in ags. was probably always (er, 
eer) or (e.r, ee.r) with a strongly trilled 
(r). It is still so in Scotland and Ire- 
land. There is no notice of its having 
varied in sound till 18., when (i) was 
recognized as a second sound of r and 
then er was taken to be {ex). In 
19. Mr. M. Bell takes it to be (aor). I 
conceive it to be properly i^S), but to 
be generally ('i), see p. 196. Although 
there is no notice of this sound in 
older writers, yet there is reason to be- 
lieve that something approaching to it 
was known in 16. and that it was well 
marked in the latter part of 17. In 
1 7. the practice of reading er as ar in : 
clerk, Derby, servant, service, Hertford, 
still more or less heard in 19. came into 
use. Confusions of er, ar, are common 
in 13. 

EU. The oldest sound of eu seems to 
have been (eu). In 14. it was generally 
(eu), but in words of French origin 
(yy), p. 302. The division became 
confused in 15., and in 16., though both 
sounds were heard, the line of distinc- 
tion seems arbitrary, see lists, p. 301. 
In the course of 17. most eu became 



(iu) though some remained (eu). In 
18. this distinction was swept away 
and all became and have remained (iu), 
except after r when they are generally 
(uu) as Jieuhen, rew, rheum. In mo- 
dern French words in eiir as : amateur, 
grand^wr, hautcwr, usage varies, (iur, 
eei, uuj, 'j) being all heard occasion- 
ally, the last being meant for the 
French (^r). [In 13. and 14. eu, tie, 
eiv, tv, each= (uu) in Norman and 
English, infra p. 586. — P.] 

EW was identical with {eu), but was 
more often used, especially in 13., and 
afterwards became the common final 
form, see eu. Some of the words in 
ew passed into {oo, oou), at least as 
early as 17., but shew, sew are in 19. 
usually spelled show, sow, and chew, 
eschetv, shrew, shrewd have (iu) or (uu). 
In Shrewsbury, present usage varies 
between (uu) and {oo). Shroiv was 
used in Shakspere's time. [See eu. — P.] 

EWE only occurs in the word (ewe), 
in 19. (jiuu) and (joo), which is found 
written aive in 13, p. 428. In the 
middle of a word ewe occurs as ew + e, 
and the e may be or may not be silent, 
as in : BeivtA, hveivedi, ]eive\ = {sood, 
bruud, dzhiu'cl). The word sewer, a 
drain, was (shoo.i) in 18., but in the 
middle of 19. the pronunciation (siuj) 
prevails. Sewer a waiter is (siu-'i), one 
who sews is (soo-'j). 

EWW. Orrmin's form of (eu), p. 488. 

EY. The same as ei, see p. 459, n. 1, 
and the passages there cited. See also 
Chap. Vil. § 1, near the beginning. It 
was common in 14. as (ai), in 16. as (ei, 
ai), in 17. as (eei, ee), in 18. and 19. 
generally {ee) sometimes {ii), as in ket/, 
they, twckey, ^^ing = (kii, 6hee, trki, 
Qviq). [See ei. — P.] 

EYE seems to occur only in eye = 
(ai), which was (ai^h-e, ai/A'e, ire) in 
14., (ei, ai) in 16., and generally (ai) 
in 17. to 19. [In 14. eye = {ee-e), in 
Norman and English, infra p. 582. — P.] 

F. In ags. (f) and between vowels 
often (v). In 13. to 19. generally (f), 
in the middle of 17- o/ became (ov) but 
it was not generally recognized till 18. 
The use of (v) for (f ) was common in 
the dialects of 14., p. 409. 

FF. Formerly in MS. of 13. and 
later ff was written for F. Through- 
out, in the middle of a word J' was = (f } . 

G. In ags. (g, gh ; y, yh, jh, j). 
In 13. a distinction was made between 
y ■^,g being pure (</, g), and 7^ guttural 
or palatal. When Fi-ench words were 
introduced more freely in 14. y became 
(dzh), and was then (d/,h) or perhaps (zh) 
in French. The sound (zh) is compara- 
tively modern in France, though it was 
certainly known in 16., p. 207, and it 
is used in Modern English words taken 
from the French as : rou^/ing (ruuzh-tq). 

GG. Identical with y, but always 
(g), never (dzh), as in rugged = (r8g-ea). 

GH. Even in 13. occasionally used 
for 3 when sounded (gh, kh), the 
sounds (^h, j) being occasionally written 
(yh, y) p. 431. In 14. the sound was 
(gh, yh, kh, kh), and after labial vowels 
(gz^h, wh). In 16. it was generally 
called (kh) but said to be lightly pro- 
nounced, and some call it (h'), others 
(wh) , and in a few words this (wh) had 
passed into (f). In other words it 
gradually became mute, in which case 
the preceding vowel had generally 
been previously altered. In 17- siyh 
drought, heiyht, were sometimes called 
(saith, drAAth, naith, and the town of 
Keiyhley is (Kiith'h') in 19. An un- 
historical h has been inserted in : ghost, 
ghastly, in which yh = {g). The (kh) 
sound is retained in : lough, (lakh), 
though it has generally become (k) as 
(lok), and as: shough, hough =(shok, 
Hak) but sometimes (naf) in groom's 
language. The change of yh into (f) 
prevailed more extensively in 17. than 
in 19., and is still heard more in the 
provinces. Varieties in 19. : GoMayhdiQ., 
hiccou^A, Bellin^/«am, hoMyh, yhost, 
lauyh, Kei^Aley = (Ksel-ansen, nik-kap, 
Bel'mdzh^m, nak, goost, laaf), besides 
being mute. Atiyh, ouyh, must be 
taken as au-\-yh, ou + yh. 

GL. Generally ^+?,but in the Italian 
word seraylio, either (Ij) or (1) from 
17. at least, 

GN. Initial, up to 16. (gn), but i^ 
17. and afterwards, the y was dropped* 
Medial, in 14. it seems to have been 
simple (n), p. 309, and this sound has 
generally remained to 19., although yn 
is incorrectly considered to lengthen 
the preceding vowel, merely because an 
e has been omitted, as in : sign, benign, 
impregn, impugn, in 14. (snn*e, be- 
mfne, zmpree-ne, empyyne), and hence 
in 16. (sain, benain-, impreen-, /mpyyn*), 
and in 19. (sain, binain*, impriin*, im- 

Chap. VI. § 1. 


z ?— J- 


piun-). In such combination as : dig- 
nity, signify, impregnate, repugnant, it 
was probably always (gn). Gill, 1621, 
acknowledges (qn) as (beniq-n), and 
some MSS. of 15. spell beningne. [In 
13. and 14. gn medial = (n) in Nor- 
man and English. — P.] 

5 ? Used extensively in 13. and 14. 
for the sounds of (gh, ^h, kh, k\ 3). 
The figure of y in the sense (j) seems 
derived from 3. The form ? being 
identical with the written form of z, 
then in use, z was also used for ^ even 
in print, see mz, z. After printing came 
into use j was soon discontinued, and 
gh, y became the usual forms. Some- 
times confused in writing with s, p. 464. 

^h used for (gh) in Orrmin, p. 488. 

H. In ags. initially, before a vowel 
(h) or (11'). Before /, r, n, w it may 
have been originally (kh), but hi, hr, 
hn, hw seem to have become (Ih, rh, 
nh, wh) in ags. times, p. 512, as they 
are in Icelandic, p. 544, and in 13. 
only (Ih, wh) remained, which were 
frequently interchanged with (1, w). 
(Wh) remains in 19., but is uncertain 
in the South. In ags, h final = (kh, 
/th). In 13. the sound of h seems to 
have been very uncertain, and in 14. it 
was lost in those words before which a 
vowel was elided. In 16. it was pro- 
nounced or not, differently from the 
present custom. In 19. it is much 
more pronounced than formerly, but in 
the provinces and among the unedu- 
cated, it is almost always lost. 

I vowel, for i consonant see ,/. In 
ags. (i, ii) or («, u). This sound seems 
to have been prevalent in 14., and the 
short value {%) lasts in 19. During 15. 
many of the words having long {ii) re- 
ceived short (^■) owing to throwing back 
the accent, but those long {ii) which 
retained the accent became (ei), and 
retained that sound in 16., changing to 
(ai) in 17., where they remain. Only 
a few modern French words have (ii), 
as invalid (inv^liid-) also called (mvsel*- 
td), in another sense. 

I A. [In 13. and 14. ia, ya, (in one 

syllable) = ai, ay = (ee) in Norman and 
English, p. 582.— P.] 

IE, medial. Occurs occasionally in 
14. as simple (ee). In 16. it was not 
much used, though it seems then to 
have been (ii) even in friend, and in 17. 
it was firmly established in a few words, 

without any historical or etymological 
reason, as (ii), and has so generally re- 
mained. In final syllables it was much 
used in 14. as (-u'-e) and in 16. as re- 
presenting the 14. final -ie, -e, and 
sometimes -y. This termination was 
generally called {-i) but sometimes (ei, 
ai). In 17. it was gradually replaced 
by y. In a few words as die, lie, etc., 
it remains with the sound (ai). [In 
13. and 14. ie (in one syllable) =ei= 
(ee) in Norman and English, infra p. 
582.— P.] 

I-E is properly identical with long 
i, which see ; but owing to a prejudice 
against ending words in v, and to the 
necessity of putting an e after g final 
to indicate the sound of (dzh), it some- 
times represented short i (^), as in 19. 
g^■ve, h've, br/dge. In modern words 
from the French it is (ii), as: antique, 
oblique, routine, machine, pique. 

lETJ is a purely French combination, 
and in 16. interchanged with eu being 
probably pronounced (eu) ; in 17. it 
was (iu), and so it has generally re- 
mained, thus lieu is (liu) or (luu), but 
hVwtenant is usually called (leften'-ent, 
or (luuten-Bnt), and Beaul/f m is (Biu'h'). 
\_Ieu, ieiv in English, hypothetically = 
eue, ewe Norman of 13., would, if 
found = (uu), infra p. 586.— P.] 

lEW. In the word view written both 
vewe and view in 16., it is a final form 
of ieu. [See ieu. — P.] 

10. [In 13 and 14. io (in one sylla- 
ble) = oi = (uu) generally, in Norman 
and English, infra p. 587. — P.] 

IE, not before a vowel, was pro- 
bably not distinctly separated from 
er even in 14. as we have both Jirst 
and ferst. In 16. and later it seems 
to have been the same as er, and in 
19. it is either ('j) or ('.<), as in: sir, 

dirt, fir. 

IU. [In 13. iu (in one syllable) = 
iw = {uu) in Norman and English, 
infra p. 586. On p. 506, n. 2, for 
(riu'le) read (ruu-le). — P.] 

J or i consonant of the 16. and 17. 
centuries in which the distinction i, j 
was not observed in writing. In 14. 
introduced for French words, and with 
the French sound (dzh) which it re- 
tains, though in France J has become 
(zh). In the Hebrew hallelujah it was 
and is read (j), but not so in other 
Hebrew words. (Maaitsh-bseqks) for 



Chap. VI. § 1. 

Marjoribanks, is an obviously recent 

K from its earliest introduction in 
the latest ags. to the present day has 
retained the same sound (k), with per- 
haps occasional unacknowledged pala- 
tisation into {k). 

KK, often used in 14. where ck 
was afterwards employed, as (k) after 
a preceding short accented vowel in a 
closed syllable. 

KN initial, in 14. to 16. and per- 
haps for some time in 17. was = (kn), 
but in 18. and 19. the (k) was dropped. 
It is, however, still pronounced in Low- 
land Scotch. In 17. Cooper con- 
sidered kn ={nh.), p. 544, n. 2. 

L from ags. to 19. = (1). The 19. 
colonel = (k J niil) is remarkable. Lis 
occasionally not pronounced, but in 
disappearing leaves an effect on the 
preceding vowel as in : talk, half, alms, 
now (tAAk, Haaf, aamz), where / seems 
to have been lost generally in 16. See al. 

LD. The I was omitted in 17. in 
could, ivould, shoidd, having been erro- 
neously introduced into the tirst, though 
heard in 16. In Gui^^ford, the d is 
usually silent. 

LE final, after a consonant, from 
16. to 19. = ('l), as: fiddle, beadle = 
(fid-'l, bii-d'l). 

LF. In alf, the I was omitted in 
16. and a became (au), which was (a a) 
in 17. and has in 19. returned to (aa). 
See al. 

LH. Occasionally used in 13., pro- 
bably for (111), a remnant of ags. hi, see 
h, but as it interchanges with I, this 
pronunciation is doubtful. 

LL. Much used as a final, and after 
a short accented vowel in a closed syl- 
lable, as (1). In compounds sometimes 
l + l, as in soulless. In Welsh words 
initially, the Englishman says (1) in 
Lloyd (Loid), Welsh (Ihhuid), but in 
i^ango//en he generally uses (thl) as 
(Thlffin-goth-len), Welsh (Lhhan-- 

LM, aim final, omitted the I in 16, 
changing (a) into (au) which became 
(aa) in 17. and m 19. has become (aa) 
with its variants, as in balm, see al. 

LN final presenting some difficulty 
in speech, one or the other letter was 
often dropped : I was omitted in Lin- 
coln, and probably in Colne in 17., n 

was omitted in kiln in 17., changes 
which remain. 

LZ. Oldformof 1j=(1j). DaMel 
in Scotland (DteI) in England (Dail'- 
zel). See p. 310, note 1. 

M, from ags. to 19. = (m). In 16. 
probably, and later, when following 
any consonant but /, r, m was ('m) as in 
chasm = (kaez-'m) although the ('m) 
was not allowed to constitute a syllable 
in verse. Some in 19 call -Im, -rm 
(-I'm, -r'm) and this was recognized 
by Bullokar in 16. 

MB final, probably omitted bm 16. 
and certainly in 17. to 19. as limb. 

MM medial only, after an accented 
short vowel = (m), from 14. at least. 

MN final = (m) probably always in 
column; and initial = (n) probably 
always in mnemonics. 

MP. Omp, which was a French 
combination, now called (oa), was in 
accented syllables in 14. = (uun), in 16. 
(oun) and 17. 19. = (aun) as in Compter; 
unaccented it was (kon) as Comptroller. 
In 19. Campbell is often (kaem-el). 
Otherwise (mp) is fully sounded as : 
camp, limp, thump. 

N. From ags. to 19. = (n). Proba- 
bly before / it fell into m, as in Banff. 

See also nc, nk, ng. 

NC. Chiefly in compounds as in-come, 
or in the termination -nee, and then = 
(nk, ns) ; but some in 19. and probably 
early, changed n into (q) before c= (k). 

ND. Generally (nd), but the d is 
sometimes mute, as in riband, hand- 
kerchief, and in the latter case the n 
becomes (q) notwithstanding the com- 
posite nature of the word = (naeq-ker- 
tsher) in 17. and (Hseq-kitshif) in 19. 

NG. The difficulty of pronouncing 
pure (n) before the gutturals (g, k), 
caused n in such cases to pass into (q) 
in the earliest times. It is difficult to 
determine before 19. whether ng was 
simply (q), or (qg) when final or medial. 
In 16. and later the 19. customs ob- 
tained, namely ng is (q) when final, 
and preserves that sound generally 
when the word is lengthened by in- 
flection, and in a few cases w^ = (qg). 
Thus : I lo«^, thou lo;?^est, longer s. 
a lo;?^ way, have all (q), but longer a., 
longest a., stronger, stron^-est have (qg), 
Compare li??^er, fi??^er, singer. When 
ng occurs before th, it is usually called 
(qk) as length, strength (leqkth, 

Chap. VI. § 1. 



streqkth) or (qqli), but many persons 
say (lenth, strenth) which Walker 
notices as an Irishism. In French 
words w^=(ndzh) from 16., some in 
19. say (nzh) but it is against analogy, 
as chsLnffe, singe, (tshemzh, smzh) for 
(tsheendzh, smdzh). Though changing 
is used, siw^^ng is employed to keep 
the word distinct from singing. Ng 
initial = (q), is only found in foreign 

NH. A Portuguese combination for 
(nj), used in 19. in ipecacuaw/ia as (n). 

NK. In one syllable = (qk), or as 
some believe (qhk) from ags. to present 
day, see ng. 

NN. After short accented vowels 
= (n) from ags. 

NZ. In a few names, the old form 
of Saxon wj, with the sound (q) as 
Me?2des = (M2q-fz), or with the sound 
(nj) as in I)e«zil = (Den"jil), see &, and 
p. 310, note 1. 

0. From ags. to 16. apparently (o, 
go), but during 15. many long o fell 
into (uu) and for some the orthography 
was changed in 16. to oo, while for 
others the o was retained, as in do, who, 
move (duu, whuu, muuv), and in 11. go 
was occasionally pronounced (guu). 
The short o also frequently represented 
(w) both in 14. and 16. In 17. the 
long sound of o in those words in which 
it had not fallen into (uu) became [oo) 
and the short either generally (a, o) or 
even (o) in case of those words where 

was (u) in 16. In 19. the long sound 
is [oo) or as some pronounce (oou) and 
even (ou), while the short sound is (o). 
Before r = (i), the long sound remains 
foo), as ore=(oo.i) although some say 
Ipox, oo'i) and even {oo-,'i) dissyllabic- 
ally, the same as ower. The short o 
before r = (i) is supposed to remain (o), 
as fork (faak), but it frequently becomes 
(aa) and the [i) is then often dropped, 
so that Lord laud theoretically (, 
lAAd) are confused as (lAAd). See pp. 
196, 245. In comic verse or, aw, are 
allowed to rhyme as in Hood's Epi- 
curean Eeminiscences of a Sentimen- 

We went to , it certainly was the 

For the next, the most blessed of 

1 remember how fondly I gazed at my 

Sitting down to a plateful oi prawns. 

never may mem'ry lose sight of that 

But still hallow the time as it ought, 
That season the ''grass" was remark- 
ably dear, 

And the peas at a guinea a quart. 

— Comic Annual, 1831, p. 171. 
See the remarks under (j), infra § 2. 
The properly short o is in 19. some- 
times prolonged before 5, / as cross off 
= (kros of, kroos oof) or (kroos oof), 
and occasionally quite (krAAS AAf). 
Possibly in 17., whole, stone were (hoI, 
ston) as these pronunciations exist in 
America, which is tinged with 17., and 
are still heard occasionally here, being 
common in Norfolk ; from (ston) ap- 
parently, or else from (stwn), comes the 
familiar (ston) as a weight. The 19. 
varieties : are go, do, women, bettor, on, 
son, woman, compter, choir, reason = 
{goo, duu, wi'm-en, bet'i, on, son, 
wum'BU, koun-ti, kwali, riiz'n). 

OA. This is found in 13. when it 
seems to have been {aa) or (««h), or 
simply (aa), pp. 467, 498, 506. It was 
hardly used afterwards, till in the latter 
part of 16., when it was introduced as 
a new sign for (oo), the form {oo) being 
appropriated to (uu). In 17. the sound 
changed to {oo) at which it has re- 
mained, with a tendency in 19. towards 
(oou, ou). In the three words : broad, 
abroad, grortt, it was = (AA) in 17., and 
still so remains, though groat is often 
called (grot), and in groats, a farina- 
ceous food for children, it is (grrts). 
It was occasionally o+a as in oasis, 
coart, coagulate. [Infra p. 586. — P.] 

CE. Used in 19. in some Latin 
words as foetus, foetid = (fii'tos, fet*/d). 

OE was uncertainly used as a final 
in 16., with the sounds of (oo) gene- 
rally, and (uu) occasionally, Levins 
1570 has : doe, foe, roe, toe, sloe, goe, 
forgoe, moe, hoe, loe (our lo !) with 
(oo), and: shooe, fordoe, vndoe (but 
doo), with (uu), but considers these 
and : bio, twoo, no, so, tho, to, vnto, 
as words " in o desinentia." In 17. oe 
was generally {oo), but was (uu) in 
shoe. In 19. we find do^;, shoi?, felloe, 
does = (doo, shun, iel-i, doz), and oe = 
o+e in: coeval, pott (ko,ii*vT;l, poo-et). 
[See oe, p. 586.— P.] 

0-E. From 16., marks o long, but 
in some words, when v is the interposed 
consonant, as : move, prove, the o was 
sounded (uu) from 16. to 19.; love, 



formerly (luuv), passing through 
(luv), became lav). In a few words 
as : hove, rove (oo) remains. Other- 
wise the sound was that of the long 
of the time. The anomaly one 
(wan) is recent ; the time of its intro- 
duction is unknown, but it was not 
before 18. Jones 1701 gives (wa^n, 
wa3ns, waenst) as curiosities, but does 
not name (wan); Buchanan 1766 has 
(wyen, wicns) also, as the correct sounds, 
but Franklin, 1768, has (wan, wans). 
The Scotch (jm, j(?n) for ane, seem to 
have been introduced about the same 
time. The old sounds were, English 
(oon), Scotch (aan). The 19. varieties 
are : hors^, cov^, move, Tolkmache, 
forehead, love, Bolingbroke, nne = 
(Hors, koov, muuv, Toel'maish, forced, 
lav, B?d-iqbrwk, wan). 

OEU. A French combination, na- 
turalized as (mi) in manoeuvre, in 19 ? 
[A combination not known in France 
until 15., represented in 13. and 14. by 
ue, eu, eo, oe=(uu). — P.] 

OH has perhaps always represented 
the exclamation (oo), although the ex- 
clamation was not always represented 
by it. 

01 is not found often enough in 13. 
to determine its sound, it was appa- 
rently (ui) in 14. in French words, 
but occasionally (ue ?), and sometimes 
(oi?); in 16. (uui, ui) and also (oi), 
in 17. the (ui) class became (ai) and 
this remains as an unrecognized vul- 
garism in hoil, pomt, etc. ; in other 
words it was (Ai) or (oi) or (oi), and 
occasionally {oi) is heard, often (AAi). 
Dialectically oi was occasionally pro- 
nounced {ii, ee) in 14., p. 450, note 2. 
The 19. varieties are : chamo/s, conno?s- 
seui', avo2rdupoise = (shaem-i, shaem-wA, 
konesjj-, aevjdiupoiz-). Choer was also 
writen quire in 17., and since then pro- 
nounced (kwair), but chorister was 
(k^^^r•^ster). Memoz'r is called (mem*- 
wo.t) in imitation of the French. And 
sometimes oi = o-\-i. [In 13. and 14. 
oi, oy = io = (uu) generally, in Norman 
and English, but very often also = (ee), 
infra p. 587.— P.] 

OL, OLL. In 16. the I being 
sounded strongly as ('1) or (\.w) de- 
veloped a (u), so that o/ became (ooul) 
in roll, toll, etc., p. 193. In 17. this 
remained or became (oul), and as such 
passed to Ireland. Even in 18., (oul) 
as well as (ooul) was heard. In 19. 
(ooul) is considered inelegant, but is 

common, and (oul) unbearable, and 
(ool) is the only recognized sound. 

00. In 13. and 14. = (oo), rare in 
13., frequent in 14. During 15. this 
sound split into (oo) and (uu) and in 
the latter part of 16., oo was appro- 
priated to (uu), Avhere it has since re- 
mained, with a few exceptions. In 
some words the (uu) became (u) and 
some of these naturally fell into (a) in 
17., as: flood, blood; others, however, 
resisted this tendency, but became (w) 
as : good, wood, stood. These changes 
remain in 19. Before k it is the cus- 
tom in Scotland to use (u) and in the 
North of England to preserve (uu), as : 
book (buk, bunk), while in the South 
the sound is fully (m) as (bi<k). In 
some words oo=oa-o, as zoology, zoo- 
phyte, Laocoon = (zool*odzh«, zoo'ofait, 

OR. There is no reason to suppose 
that this was difi'erent from (oor, or) in 
accented syllables ; finals were gene- 
rally written our up to 17. and even 
later, some still remaining, originally 
to indicate the sound (uur, ur) p. 304. 
In 17. these final unaccented or, our 
became (ar) or probably (ai, i), and 
are (a) in 19. In accented syllables, in 
17. it was sometimes (oor) and some- 
times (Ar) or (AAr), (foorm) a bench, 
(fAArm) a shape, and this distinction 
remained through 18. It has nearly 
disappeared in 19. The present theo- 
retical sound of or not followed by a 
vowel is (oj), which passes into (aai) 
and (aa) simply, see the citation in o, 
p. 575. Before a vowel or = (or). 

OU was introduced at the close 
of 13. and beginning of 14. for (uu) 
and so remained to 16,, being occa- 
sionally used for (u), and occasionally 
for (oou), which was generally 
written ow. Some writers pronounced 
it (uu) till past the middle of 16., 
but about that time the general pro- 
nunciation had become (ou), some 
words only remaining (uu) or (u). 
Most of the latter became (a) in 17., 
but some (uu, u) remain to 19. The 
ao:s. words in aw, ow, which came to be 
written ou, ow, were till 17. called 
(oou). In 17., {oo) without an after- 
sound of (u), was and still is the recog- 
nized pronunciation, but as the after- 
sound exists still as (oolu, oo'w), it pro- 
bably existed in 17., and its repudiation 
by orthoepists then arose very possibly 
fi'om the same cause that it still arises, 

Chap. VI. H- 



namely, the tendency to give this after- 
sound (lu) even in words where there 
is no historical authority for its use, 
see ow. Before gh the sound was ap- 
parently (ou) or (oou) in 14. In 17. 
this changed to (aa), gh being dropped, 
and has so remained. The 19. varieties 
are : o«<ght, sowl, so?<p, hough, &ou\)\e, 
wo?/ld, noim = (AAt, sool, suup, Hok, 
dab*'!, w^^d, naun), and it is sometimes 
o+u. [In 13. and 14. om=(uu) in 
Norman and English. — P.] 

OUGH, properly = 02« + ^/i, and its 
noted varieties arise from the combina- 
tion of the varieties of these two sym- 
bols, which they do not exhaust. In 
19. they are : though, totcgh, hiccoi{gh, 
"plough, through, lough, hough, ought = 
(dhoo, taf, Hik-kap, plan, thruu, lakh, 
Hak, AAt). These are only eight ; 
as there are at least seven varieties of 
ou and of gh, ough might have had 49 
sounds. It is not the combination of 
the most varied pronunciation, as is 
generally supposed, for simple o has at 
least 10, and eo 11 uses, see o, eo. 

OW in 14. was generally used for 
(oou), but sometimes was written for 
ou and pronounced (uu, u). In 16. 
those words which had (oou) retained 
the sound. In 17. they changed (oou) 
into {oo) which remains. There is a 
strong tendency to say (oon) in 19., and 
as this tendency is as strong for no as for 
k>wtv, orthoepists disapprove of it in 
both cases, p. 234. Those words in which 
ow was called (uu) in 14., were pro- 
nounced with (ou) in 16., and (au) in 
17., which remains as hoiv, noiv. The 
19. varieties are : knoi^, Cotvper, kno?^- 
ledge, bellflzf;s, now = (noo, Kuu-p.i, 
nal'//dzh, bel as, nau). Coivper is some- 
times called (Kaup-.i). [In 13. and 14. 
ow generally = (uu) in Norman and 
English, and sometimes (oou) in Eng- 
lish.— P.] 

OY can only be regarded as another 
form of oi from 14. to 19. It is now 
generally final. [In 13. and 14 oy = 
(uu) generally, but often = (ee) in Nor- 
man and English, infra p. 587. — P.] 

P. From ags. to 19. =fp). In cu;?- 
board it is in 19. assimilated to the 
following b, or rather lost= (kab-jd). 

PII was introduced at the earliest 
periods lor Greek c^, and probably 
always = (f). In ne;;//ew the ph was 
a mistake, and it is called (neviu) in 
19, In Cla/?//am, etc., ph—p-^h and 
the h is dropped (Klcep-Bm). 8eephih. 

PHTH, properly ph+th, is only used 
in Greek combinations. From the dif- 
ficulty of saying (fth), the following 
changes arise : phthisis, pht/dsica,l, apo- 
phthegm, diphthong = (tBi-S''s, tizikTBl, 
sep'othem, d/p-thaq). The last at any 
rate was in use in 17. We find even 
in ags. (pth, kth) used for (p9, x^ ^ 
transliterating Greek, p. 523. Some 
say (difthaq) in 19. 

PN initial loses p, as in pneumatics 
= (nium8et-iks). 

PP after short accented vowels = (p). 

PPH after short accented vowels 

= (f)- 

QU from 14. to 19. had the sound 
(kw) or (kw) . In a few words from 
the French it is (k). These were for- 
merly spelled without git, compare 14. 
licour, 19. liquor = (likuur-, hk-j). 

QUH. An old Scotch orthography, 
probably representing (ktt"h),the Scotch 
substitute for English (wh). 

E. From ags. to 19. before a vowel 
= (r), and perhaps once (.r). In Scot- 
land always (r) or (.r) wherever occur- 
ring. There is no mention of any such 
sound as (a, .«) till 19., but there is 
reason to think (.i) may have existed in 
16. and still more that it existed in 17. 
For its use in 19. see table on p. 197. 
There are many varieties of defective 
utterance. The Northumberland burr 
is (r) or (grh) and sometimes (gh, g) 
simply, the French r grasseye ou pro- 
ven(^al is (/•), and the Dutch g ch have 
often the same sound, thus schip = (s/v p) . 

EE final, seems to have been occa- 
sionally (er) in 14., but when the e was 
inflectional (re) remained. In 16. and 
later it was always (er, ur) or (jc) in 
French words. 

EH initial in Greek words and in 
Ehine, llhone = (r). 

EE Generally after a short vowe 
= (r), and possibly always so before 17" 
In 19. it is generally (r) after a short 
vowel, except there is acknowledged 
inflection, and then it is (.ir), but after 
a long vowel it is always (jr). Thus: 
marry, merry, spi>it, horrid, hurry = 
(mssr-i, mer-/, sp/r<t, Horwd, nar-i). 
But occu'-, occurrence, occurring, infer- 
ring =(akj-, akar-'Bns, okjr^q, infr, 
infj r/q) . After a long vowel rr is 
seldom written, the single )• being then 
pronounced as (ir), compare: earring, 
hea'ing = (iirr/q, niiir/q). But we 
have : tar, tarry = covered with tar, 



VALUE OF LETTERS. RRH — p. Chap. VI. ^ 1. 

star, starry = full of stars, = (taaj,taaaTi 
staaa, staaa-i^), and in Ireland arr 
always = (aaar) or (a^ajr) as in barrel in 
England (ba,'r-el), in Ireland (baa.iTel) 
or (bieic'rel), which seems to imply a 
similar English pronunciation in 17. 

HUH, in words from the Greek 
only, in 19. used precisely as r, rr, as in 
catarr/i, diar/7/a^a=(k^taai', daiurii-B). 

S. One sense of this letter from ags. 
to 19. has always been (s). Whether 
in ags. it was ever (z) is difficult to 
determine. Judging from the Ice- 
landic, as the representative of medieval 
languages, s was always intentionally 
(s) in ags ; but the sound of (z) was oc- 
casionally generated. Kapp takes it to 
have been always (sj). This is not 
necessary. There is no (z) in Spanish, 
nor in the Dyak languages, and pro- 
bably many others. In 14. there 
seems no doubt that s was occasionally 
(z). There are some traces of its being 
changed into (sh) by a following pa- 
latal vowel at the ead of 16. and be- 
ginning of 17. (p. 215), and later on 
in 17. Miege, a Frenchman, notes : 
sure, leisure, usual, as being (shyyr, 
lee-zhor, yyzh-ya^l). See sci- si. These 
sounds remain. In 19. we have: see, 
a*, sugar, lemire = (sii, sez, shug-i, 
lezhj). In some MS. of 13., st is 
used for ^t = (7cht), probably a mistake 
arising from the confusion of ^, j, z, 
see p. 464. [In 13. and 14. s = (s) in 
Norman and English. — P.] 

SC. The initial sc before palatals 
was (s) in 16., and probably always. 
A^ceptic was often spelled skeptic. In 19. 
we have : viscount, scene, discern, sceptic 
= (v8i'k8unt, siin, dizin*, skeptek). 

SCH, in Greek words, seems to have 
been considered as sk (sk) . The words : 
«cAism, sc7/edule, have always presented 
difficulties. They are now generally 
(s^■z•'m, shed'iul). In 13. and 14., and 
even later, sch was used for the mo- 
dern sh, which see. In 13. it is some- 
times she. The celebrated German 
name of Rothschild, properly (Root'- 
%\i\\(i) =red-shield, is generally mispro- 
nounced in English as (roths-tshaild), 
quasi Wroth'' s child! where the familiar 
word child has evidently misled the 
reader to separate the combination sch. 

SCI-. Treated as si- = {^i) till 17., 
and then often (sh), as in 19., consmus 
= (kon-sh'Bs), 

SH. Orrmin uses this compendious 
form of sch^ but it did not come into 

general use till end of 15., or beginning 
of 16. It represented the effect of pala- 
tizing (sk), and henco converting it 
into (sh). The sound (sh) has re- 
mained. Sh is occasionally s-^ h, and 
the h is occasionally dropped, as 19,, 
compare misAap, disAonest, dishonour, 
MasAara = (m«sHa;p", d/son"est, dtzonu, 
Majs-Bm) ; but many persons ignore 
the composition, and call : Hors/jam, 
WindlesAam (Ho.ish*Bm, Win'd'lshijm). 
The pronunciation (thresh -Hoold) for 
thres/iold, ags. )?resc-wald, Chaucer 
threisshfold, 3482, Promptorium 
threschwolde, is a modem etymological 
error for ( thresh oold). 

SI-. Treated as (st) till 17, and 
then often (sh), and sometimes (zh), as 
19., mans2on, decision = (msen-sh^n, 
disizh-en). After a short accented 
vowel it is more usually (zh), and (sh) 
is then kept rather for ci-, or ssi. 

SS was occasionally used for (sh) in 
13. and 14. (pp. 409, 448). 

SSI- See si. 

T from ags. to 19. =(t); but see ti-. 

TCH intended as double eh, and 
used after a short accented vowel ; the 
spelling is modern, the 14. form is cch. 
In both cases the sound was probably 
(tsh) simply. 

TH, even in ags. used as a trans- 
literation of Q, p. 523, and sometimes 
used for }?, "S, in 13., having both the 
sounds (th, dh), which were probably 
distinguished as at present in 16., with 
some doubtful cases, as wi^A (weth, 
widh). Sometimes = t-\-h, sometimes 
t + th, or th + h, being obviously con- 
tractions. In a few words th = (t, d) in 
16. In 19. we find : thjme, bur^Aen 
(generally written bur^^en), ^Aigh, thy^ 
poMouse, eighth, Sou^Aampton = (taim, 
bid'n, thai, dhai, pot-H8us, cetth, 
Sauthnajm'tBn). In Havelock th is 
found for jt, as Jcnith, but the sound is 
unknown ; it may have even been really 
(th), compare si(/h, Keighley, under Gh, 
or else simply (t), p. 477. 

TI. In the termination -tion, pro- 
bably (s?) from 14. to 17., and then 
generally (sh), following si-, ci-, sci-. 
It may, however, have been exception- 
ally (sh) even at the beginning of 17. 

TTH, the Greek t9, probably al- 
ways (th) in Ma/^Aew. 

p ags. (th) or (dh). It is impossible 
to distinguish between j? '5 in ags. and 
Early English. In 13. and 14. used 
for both (th, dh). In ags. it is safest 

Chap. VI. § 1. 


u — uy. 


to use (th) initial, and (dh) medial and 
final, p. 515 and p. 541, n. 2. 

U vowel, for u consonant, see v. In 
i^s. (uu, u). In 13. the long u was 
(an), but may have been occasionally 
pronounced (yy) likewise, while short 
M, though generally (u), was occasion- 
ally either (y), or (i, e). This usage of 
short u is too general to be considered 
as dialectic. In 14. long u was always 
(yy), the (uu) sound being represented 
by ou, ow, which see. Short u was 
more uniformly (u), though this sound 
was occasionally written ou, as the use 
of short u for (i, e) had not died out. 
In 19. this use of short u is only re- 
tained in: bm-ial, bwry ; bwsy, b?«siness. 
In 16. long u was (yy), and short u (u) 
almost uniformly. In the beginning of 
17., and perhaps earlier (p. 227, n. 1), 
long ti was called (yy) by some, and 
(iu) by others, the latter sound pre- 
vailed, and has remained to 19., except 
after r, as in trwth, rwle, and after an 
s palatalized into (sh, zh), as : sure, 
leisure, when it becomes (uu), or is 
lost in 19. as : (truuth, ruul, shuuj, 
lezhu). There is, however, great di- 
versity of practice, and an (i) is more 
or less distinctly introduced before the 
(u), as (iu, iu), or fused with it in (yy, 
uu). Again, in the middle of 17. short 
u became generally (o), which was a 
new sound in our language, not men- 
tioned by any writer before Wallis, 
1653, and the extent to which it was 
used is very undefined ; but it prevailed 
generally, and only a few (u) remain in 
19. which are now properly (w), as : 
pwt, f «11 = {put, ful) . This uncertainty 
is well illustrated by the dialects of the 
peak of Derbyshire, chap. XI., § 4. In 
16. short u was occasionally called (^), 
but this was reckoned an affected pro- 
nunciation. The use of u for w in 
persuade, etc., is modern, imitated from 
its use in qu. In 16. or 17. arose the 
practice of using (/u to represent a hard 
ff (g) before an e, as iu ffuess, a French 
practice, borrowed also from qu ; and to 
this, and the wish to indicate a long 
vowel by final e, must be attributed 
plague, vague, fatigue, rogue, etc. 
"With usual inconsistency a long vowel 
is not always indicated by a final -gue, 
as epilogue, synagogue^ or tongue. 
These spellings are not found before 
16., and they greatly vary in 16. [In 
13. and 14. « accented and long = (uu) 
in Norman and English ; u unaccented 
and short = ('B, e, i), and u with the 

secondary accent = (a, e, t), infr^ p. 
583.— P.] 

UE used in later spelling as a final 
u, owing to a rule made by no one 
knows whom, no one knows why, and 
no one knows when, that no English 
word can end in u. [In 13. and 14. 
ue = eu = w = (uu) in Norman and 
English, p. 586.— P.] 

U — E from 16. indicated long «, and 
was so pronounced, see u. 

UI. This is not properly an Eng- 
lish form, but it is found rarely in 14. 
in place of oi, with, probably, the 
sound (ui). In some words it may 
have been (yy), as in them it often in- 
terchanges with simple u, p. 135 and 
170. See also p. 424, note 3. Some- 
times it replaced i, see p. 452, note, 
col. 2, 1. 8. To this custom is perhaps 
due its present existence in bwild, which 
Gill 1621 calls (byyld, boild, biild, bild), 
and which is spelled beeld, hild in 
Promptorium. After g the u was only 
the French method of hardening g to 
(g) and the combination gui must be 
considered as g hard + «', as : guilt, 
guide, guile. In more recent 17. 
French words, ui was treated as long 
u, and this treatment remains with the 
sound (uu) after r as usual, and some- 
times after s, as s««t, 17. (suut), 18. 
(shunt), 19. (siut). Oecasionally ui = 
IV + i, or = w + i. Hence we get the 
19. varieties : mosqt«to, ixuit, b««'ld, 
gw/ding, &uii, lang«<«d, quirk, frm"tion, 
angw/sh = (moskii-to, fruut, bz'ld, gaid*- 
i'q, siut, laeq'gie;2'd, kM;erk, fruesh"yn, 
ee•giu,^■sh.) It is continually used in 
Scotch for (yy) or (a) as : puir^ guid. 
[In 13. and 14. ui = uy = iu = (^\i) in 
Norman and English, infra p. 586.-P.] 

UO. [In 13. and 14. mo=om = (uu), 
when u is not a consonant, in Norman 
and English. — P.] 

UOY is confined to the word buoy, 
called by Hart 1569, (buee) = (bwee), 
in 17. (boi), frequently (bwoi) and by 
sailors (buui) in 19. 

XJR, from the time that u short re- 
presented {Q),ur = {dx, 9j, 'r, j), see p. 
200, er and r. 

UW, an unusual and hence doubtful 
combination, probably (yy). [In 13. 
and 14. uw = (uu) in Norman and 
English, infra p. 586.— P.] 

UY, a modern spelling, found in: 
hut/, plag?^?/ = (bai, pW-g«). The sound 
of buy, spelled: bye, beye, 14. was 
(bu-e, bai-e), p. 285. [In 13. and 14. 



V— Z, 

Chap. VI. § 1. 

uy = ui = hi = (uu) in Norman and 
English, infra, p. 586.— P.] 

V consonant, for v vowel see u. This 
seems to have been invariably (v). 

W vowel, is only used as part of a 
diphthong, see aiv, ew, ow. Several 
writers, however, consider w to be 
always a vowel. In 13. occasionally 
used as long t<=:(uu), especially where 
(uu) dialectically replaces (wuu, wu) ; 
in 14. occasionally used as ou also = 
(uu) ; probably double v was dialectic- 
ally used as the simple v vowel, that is 
«, with its local sound (uu) or (yy). 
[In 13. and 14. tv=ew = u = {uu) in 
Norman and English, infra p. 586. 

W consonant, corresponds to ags. p, 
which was (w) p. 513. This sound 
has remain to 19. ; and is often con- 
sidered to be a vowel, but it is not so, 
compare m'OO, wood, w^oman = (wuu, 
wud, wura-Bn),in which those who con- 
sider w as a vowel have to write (uu, 
ud, um'tjn), as is and probably was 
frequently said in various parts. Mute 
in 19. in: gunwale, boatsirain, anstrer, 
Chismck, st^ord, two, ti^opence ; the 
last word was (top-ms) in 17. In ags. 
p. 514, and down to 16. at least wr- 
initial was probably a labial r or {tw) 
as write, (ureit) in Hart, (wrait) in 
Gill, but simple (rait) in 19. Ags. tvl-, 
p. 514, was probably a labial I or (1?^;), 
which changed to (1) or (fl), compare 
ags. wlaenco, Scotch wlonk, modern 
flunkey ; Is Iwkewarm a transposition 
of ags. wlxc ? Orrmin has wlite. 

WH, in ags. hw, was perhaps very 
early = (kMh), but is not likely to have 
been (khw). In Scotland it is assumed 
as (ku-'h,) see quh. Probably in later 
ags. times it was (wh) and it has since 
80 remained, though there was a ten- 
dency even in 13. to call it (w) when 
initial, and that tendency is strong in 
the South in 19. In 16. who was called 
(whuu), which in 17. had become (huu) 
where it remains, (whoo, whuu) being 
heard from elderly provincials. The 
final wh in 14. formed the transition 
from (ki^-h) to (f), and in Aberdeen 
(fat) is still said for (kw;hat) quhnt, 
what, the same transformation occur- 
ring initially. 

WL. See w. 

WR. Seew. 

X was in early writings used for 
Greek x ^^ XpiaTos, whence the con- 
tractions Xp' = Xp. Xmas, etc., for 
Christ, Christmas, etc. ; and was then 

= (k). Its generaF early use was for 
Latin x, and it seems to have been 
always (ks) and never (gz). In 19. it 
is sometimes (gx), and being treated aa 
k+s, or ff+z, the latter letter may be 
palatalized to sh, zh. In French words 
it follows the French pronunciation 
(s, z), and as an initial in Greek words 
as pronounced in English it was (s) in 
17. and is (z) in 19., as Xantippe, 
Xenophon, Xerxes, now = (Za)nt?p*i, 
Zen-Bli;n, Zerk'ziiz). Hence the 19. 
varieties : earcept, beau:r, ycz, aadom, 
earample = (eksept-,booz, veks, ak-shi^m, 
egzaam p'i). [In 13. and 14. x = {s) 
in Norman, and often perhaps in Eng- 
lish.— P.]. 

Y vowel, was in earlier ags. (y, yy), 
but in later ags. times it was confused 
with (i, ii). In 13. to 16. it was used 
indiscriminately with i, as of precisely 
the same meaning. In 17. and subse- 
quently the use of y was more limited 
to the end of words, where it arose 
from the termination -ij, the y being 
in 14. the substitute for 3, in this sense, 
and the i omitted. Throughout, the 
Latin practice of transliterating Greek 
u by y was followed. The pronuncia- 
tion of 7/ vowel was the same as i vowel 
throughout, see i. In 19. compare 
marry, myrrh, flying = {msur'i, mi, 

Y consonant. This was a substitute 
for ags. ^, and its use probably arose 
from the sound of 3 as (j). It has 
been used for (j) from 14. at least. It 
was also used in contractions for ]?, as 
ye yt = J,e >8et. 

YA. [In 13. and 14. pa (in one 
syllable) = ai/ = ai={ee), in English 
and Norman, infra p. 582. — P.] 

YE. [In 13. and 14. ye (in one 
syllable) =ey in medial, and sometimes 
probably in final syllables = (ee), in 
Norman and English, infra p. 582. — P.] 

YH. This is found in 13. in place of 
3 when it had the sound of (j), p. 431. 

Z is not an ags. letter. In 14. it 
was freely used for (3) even in plurals, 
see Alliterative Poems, edited by R. 
Morris, and also for 3, and had there- 
fore both sounds. The use of z for 3 
remained into Roman type, see 3 and s. 
In 16. its use was contined to iz), and 
it was abandoned m plurals. In 19 it 
is palatalized and a few Italian z's are 
found, hence : mezzotint, zeal, azure = 
(met-sot/nt, ziil, ee-7\ii). [In 13. and 
14. z, zs = (s), in Nor. and Eng., and 
sometimes perhaps (ts) in Norman.-P,] 

Chap. VI. § 1. MR. PAYNE's HESEARCHES. 581 

Having learned that Mr. Payne in the course of his I^orman in- 
vestigations (supra p. 438, n. 1) had arrived at several results 
which were inconsistent with the preceding investigations, I re- 
quested him to give me that brief statement of his opinions which 
has been added in brackets to several of the above articles, and also 
to furnish an abstract of the grounds on which he relied. This he 
has been so kind as to do, and it seemed to me so important that 
the reader should be in possession of his arguments, that I have 
here appended them in extenso. In his Memoir, above referred to, 
the several points here shortly touched upon will be fully illustrated 
by citations and references. It would be impossible fully and 
satisfactorily to criticise his investigations without studying those 
additions. At present I can only add brief notes, pointing out the 
radical difference between our views, which, as respects ai/, ey and 
long w, will be further illustrated at the beginning of Chap. VII. 
§ 1, and state my opinion that, as far as English is concerned, suf- 
ficient weight has not been given by Mr. Payne to the dialectic 
peculiarities of the scribes of MSS. Thus it appears to me that 
the Alliterative Poems in the West Midland dialect of the xivth 
century, afford no proper evidence for Chaucer's pronunciation in 
the South, and the late xvth century MSS. of Alisaunder used by 
Weber (supra p. 451, note, col. 2) is no authority at all for the 
pronunciation of the xiiith century to which the original poem 
belonged. The assumption that so many forms were used to express 
the same sound, so that the vowels (uu, ee) must on this theory 
have been predominant in the English and Norman of the xm th 
and XIV th centuries, seems also incompatible with the known ten- 
dency of all illiterate speech to diversity of pronunciation. Thus 
&tone was ags. (stflf^n), and is in ordinary Scotch (st^m), but in 
Aberdeen (stiin), in Cumberland and Westmoreland is dubiously 
(stjaan, stii'aan, stii'^n), in the xvith century probably (stoon) as 
it now is frequently in the provinces, in the xviith century and 
still theoretically (stoon), but probably often in xvnth century, as 
it still is in Norfolk and the United States (ston), whence the com- 
mon form (stan) for the weight, and perhaps the most usual em- 
phatic southern pronunciation is (stooun). Such diversities in olden 
times must have produced diversities of spelling. See also supra 
p. 473, note, col. 2, for {ee^ oi). I take this opportunity of pointing 
out the necessary deficiencies of my own investigations upon 
English pronunciation during the xni th century, which ought to 
have been based upon an extensive examination of existent English 
dialects, and a thorough comparison of the various MSS. of the 
same works written by scribes in different parts of the country, 
as checked by the knowledge thus gained of their local peculiarities. 
Had I waited until this was possible my book would probably never 
have been written, and the circumstances under which this part of 
it was unavoidably composed did not even leave time to undertake 
so thorough an examination as I could have wished of all existing 
documents and sources of information. The reader is therefore 
requested to consider Chap. Y. rather as the commencement than 


MR. PAYNE ON AE, EA, AI, lA, EI, IE. Chap. VI. § 1. 

tte completion of a research, whicli the labours of such competent 
investigators as Mr. Murray for the Scotch dialects, Mr. Sweet fof 
the Northern languages, and Mr. Payne for the Norman element, 
will contribute to advance, but which may require many years of 
patient study both of existent and extinct dialectic usages, not only 
in England, but low Germany and Normandy, to bring to » 
thoroughly satisfactory conclusion. 

The remainder of the text of this § is written by Mr. Payne ; the 
footnotes are by myself, but have been signed for greater distinctness. 

Brief Abstract of some of Mr, Payne's Researches on the Valttb of 
THE Letters in Norman and English, during the Thirteenth an© 
Fourteenth Centuries. 

AE, EA, AI, lA (in one syllable), EI, IE (in one syllable), with thi 
variants AY, YA, EY, YE = (ee). 

Assumino: the Norman Ions: or tonic 

e to have been = (ee), and finding it in 
Norman poems of 13. frequently rhym- 
ing with ei, ai, as : feel conseil, defens 
mains, estre maistre, nestre maistre, 
fere plaire, retraire manere, brait set, 
plein foren, reis Engles, reis pes =paix, 
consail vessel, reis lees = /o/.s, jammes 
curteis, feiz turnez past participle, re- 
fait De, etc., etc., and finding also : 
faire fere, maistre mestre, aveir aver, 
conrai conrei conre, trait treit tret, 
etc., etc., continually interchangeable 
with each other, we can scarcely help 
concluding that Norman ai, ei={ee).^ 
We infer then that pais of the Saxon 
Chronicle and Layamon, pays of Robert 
of Gloucester, payse of Dan Michel, 
were (pees), and this inference is con- 
firmed by finding the ai, ay, translated 
into e, eein pes of Owl and Nightingale, 
pees of Piers Plowman and Cbaucer,^ 
whether these be considered as literal 
adaptations of the Norman form (see 
above), or phonetic representations of 
the English ai. On the one hypothesis 
the Norman ai seems to be established 
as (ee), and the N ormsin faile, fai, crei, 
which are found rhyming respectively 
with English taile, dai, aivey, must 
have been (feel*e, fee, cree) ; and if so 

' See cause for doubting the generality 
of this conclusion, suprtl pp. 454-459.— A.J. E. 

8 This point is considered in Chap. VII. 
} 1, near the beginning.— A.J. E. 

8 For evidence that day, way were not so 
pronounced, see the table p. 489.— A.J.E. 

* This is also Rapp's hypothesis, but to 
me the origin and progress of the orthogra- 
phy appears to have been entirely different. 
Suprk p. 425, and infra p. 588, n. 4.— A.J.E. 

* West Midland, and hence of no autho- 
rity here. See supri p, 451, n. c. 1.— A.J.E. 

it is difficult to see how the English^ 
words could have been other thai 
(teel-e, dee, awee-).^ On the othei( 
hypothesis ee represents, at the will 
the writer, English ai, and, therefore, 
the Norman and English phonetic 
systems being by hypothesis the same,* 
English ay, ey, would, correspondingly, 
represent Norman e, ee. And this we 
find to have been the case. The Nor- 
man word jornee or j'urnee, became in 
Genesis and Exodus iurne, which ia 
the Alliterative Poems is journay,^ 
and in Mandevile journei,^ probably 
pronounced (dzh'ernee-) . The English 
ay is here obviously employed to re- 
present the Norman ee. The word 
contrey in Alisaunder,'' contraye in Dan 
Michel, 8 similarly represents Norman 
cimtre or contree, and in regard to both 
words it is difficult to see how the fact 
that the English ay, ey = {ee), could 
have been more clearly expressed.' 
The ay, ey, being no part of the Nor- 
man word, would appear to have been 
chosen as suitable phonetic equivalents 
to the Norman ee. These words con^ 
trey, contray, jornay, rhyme in their 
turn with Norman fey, fay, and thus 
shew that the Norman ai, ei, were also 
= (ee). The general argument is con- 

6 There is no contemporary MS. authority 
for Mandevile.— A.J.E. 

7 A discredited MS. for this purpose, 
supra p. 451, note, col. 2.— A.J.E. 

8 Dan Michel's use of ay is considered in 
Chap, VII. \ 1, near the beginning. There 
is no reason to suppose that such an inde- 
pendent orthographer was guilty of such % 
solecism as to use ay and e indifferently. — 

9 There is a great accumulation of evidence 
on the other side, already given in this 
work.— A.J.E. 

Chap. VI. § 1. 



firmed by the rhymes : maide misrede, 
maide grede, in Owl and Nightingale, 
and : maide muchelhede in Floris and 
Blancheflur (E.E.T.S. ed. p. 52)/ 
which form a parallel to : retraire fere, 
maistre nestre, etc. in Norman. We 
conclude then that ai, ay, ei, ey, 
whether Norman or English was in 
13. and 14. =(ee).2 This sound may 
have persisted generally, therefore, 
to 15. also, but in 16. Mr. Ellis's 
authorities and arguments (supra pp. 
118-124) seem to prove that it was for 
the most part superseded by (ai), though 

the old pronunciation was probably still 
extensively used.^ But the sound (ee) 
had other graphic representations. On 
the hj'pothesis, which there seems much 
reason for adopting, that both in Nor- 
man and Early English the transposi- 
tion of the vowels of the digraph, 
made no difference in the sound, ae, m, 
ai, ia (in one syllable), ei, ie (in one 
syllable), with their variants ay, ya, 
etc. would all = (ee). There is, how- 
ever, no adequate space here to illus- 
trate this position. 

AU = (au) AND (§.a8) or (aa«). 

As au in Latin was most probably 
pronounced (au), there seems every 
reason to believe that the initial and 
medial au was the same in Norman. 
This is confirmed by a remark of Beza's 
(supra p. 143, note), who especially 
distinguishes the Norman pronunciation 
of au from the ordinary French, telling 
us that in Normandy in 16., autant was 
pronounced nearly — perinde pene acsi 
scriptum esset — a-o-tant^ This pro- 
nunciation is also, I believe, still heard 
in some parts of Normandy. The old 
spellings Awwstin for Austin (supra 
p. 489) fawte faute, maugre maugre, 
hawte haute, hawnteyne, corruption of 
haultain ?, pawtenere pautoniere, etc. 
seem to confirm this notion. In the 
case, however, of the termination — 
-aunce, found not earlier than 14., and 

then taking the place of a previous 
-ance, there is much reason to doubt 
whether the rule applies. ^ The u is 
evidently not organic. It seems to be 
merely intended to lengthen out the 
sound of the a, and thus emphasise 
more strongly the accented syllable. It 
is most unlikely that a sound which 
had been established for ages as (aa), 
should suddenly change to one so 
different as (au).^ This view is con- 
firmed by the fact that in Anglo-Nor- 
man texts — it is found in no other — 
ance very frequently rhymes with amice. 
The same remarks apply, mutatis mu- 
tandis, to such words as graunt granter, 
haunt banter, commaiind commander, 
etc., which were most probably pro- 
nounced (graagnt, naaant, komaaand*),"^ 
if indeed the u was really sounded at aU. 

U Long, Tonic = (uu). U Short, Atonic = (^, o, e, ^). 

If the medieval Latin long u was 
(uu), which is generally acknowledged, ^ 
it is diflScult to see how the Norman 
long u, which often rhymed with it, 

as : la sus equinoctius, juggium con- 
jugium, etc., could have been anything 
else. If, however, it is objected that 
these Latin terminations are not long, 

1 These are considered in Chap. VII, \ I, 
near the beginning'. — A.J.E. 

2 The evidence here, necessarily imper- 
fectly, adduced, does not incline me to 
change the opinions heretofore expressed, 
of which corroboration is afforded by an 
examination of the usages in seven MSS. of 
Chaucer's Prologue and Knightes Tale, in 
Chap. VII. \ 1. See also p. 459, n. 1.— A.J.E. 

s This hypothesis seems to me incon- 
sistent with the general custom of the change 
of pronunciation. The change of (ai) into 
(ee) is common, p. 238, and could not but 
have proceeded vpith different velocities in 
different countries and parts of the same 
country. — A.J.E. 

* Beza, as quoted by Diez, also says p. 41, 
"majores nostri — sic efferebant ut a et i, 
raptim tamen et uno vocis tractu prolatam, 
quomodo efferimus intcrjectionem mcitantis 
hai, hat, non dissyllabam, ut in participio 
hai (exosus), sed ut monosyllabam, sicut 

Picardi interiores hodie quoque banc vocem 
aimer pronuntiant." The histories of ay, 
aw are parallel. — A.J.E. 

6 See the quotations from Palsgrave and 
Salesbury, supra pp. 143 and 190, for the 
reality of (au). — A.J.E. 

6 There is no change of the vowel, merely 
the insertion of a new vowel, which did not 
produce a labialisation of the first element 
for more than 200 years.— A.J.E. 

"> This almost agrees with Bullokar's 
views. — A.J.E. 

8 It is no more likely that different 
countries should have pronounced the Latin 
u alike in the middle ages, than at present. 
The French may then, as now, have called 
it (yy), supra p. 246, 1. 27. It was (yy) in 
England in 16. See infra p. 586, n. 5^ for 
remarks on the provincial character of the 
Alliterative Poems and Sir Oawayne. — 



Chap. VI. § 1. 

the answer is, that they are long as 
being under the accent, so that -us, 
-urn, would be (-uus, -uum).' Applying 
this test to English we should treat the 
us in English t/ms (C. T. v. 13384) and 
the -us in iynotius, which rhymes with 
it, as both long, and = (uus). If then 
the Norman ti was = (uu), as most of 
the authorities allow, though some of 
them speak of exceptions which they 
do not cite,'^ adventure, quoted on p. 
298, would have been (adventuur-e) 
and lure, with which it rhymes, (luur-e), 
and nature (natuur-e). (See nature 
written natwre in Alliterative Foems, 
p. 59, and salue rhyming to remwe in 
Sir Gawayne, p. 47). There appears 
indeed no proof whatever that the 
French (yy) was known in 13. and 
14., but there are many proofs that u 
was consistently (uu).^ But as it is 
generally allowed that the English 
or Anglosaxon long u of those times, 
with which the Norman is continually 
found rhyming, was (uu), proofs are 
scarcely necessary.* The greater diffi- 
culty lies in proving that the short u, 
or unaccented //, was not (u, w), but a 
different sound, approaching, if not 
identical with the obscure sound heard 
in the atonic am a man, e in the man, 
in to-day, and represented generally 
in palaeot}^e by (t') or (a, e, e, i). 
It is highly probable that this sound 
scarcely, if at all, differed from the 
atonic e of the French le in le livre, and 
that, in time, it generated the proper 
French eu. The development of this 
doctrine is essentially connected with a 
true conception of French, or, as far as 
we are concerned, the Norman system 
of accentuation. The Norman dialect, 
— and the remark applies equally to 

1 That the accent lengthens the vowel on 
which it falls, is a phonetic theory which 
has been long since abandoned. See supra 
p. 556, n. I.— A.J.E. 

2 But see supra p. 424, and especially the 
latter part of note 3. — A.J.E. 

8 That English u in 14. was (uu) and not 
(yy) seems inconsistent with the double or- 
thography u, ou. See supra pp. 298, 303, and 
infra Chap. VII, \ 1, near the beginning. 
See also p. 583. n. 8.— A.J.E. 

♦ It seems to result from my investigations 
in Chap. V. that u ceased to represent (uu) 
in English during the period 1280 to 1310, 
when ou was gradually introduced as the 
representative of that sound. See especially 
p. 471, n. 2.— I don't know to what other 
writers Mr. Payne alludes.— A.J.E. 

^ Direct proof would be necessary to es- 
tablish this remarkable difference between 

the actual Norman patois, — seems to 
have been characterised by an extremely 
strong and emphatic delivery of the 
accented syllable. The general prin- 
ciple of the accentuation consisted in 
singling out for the tonic accent the 
syllable which was accented in the 
Latin original, so that, for instance, 
Norman raisun from ration-em was ac- 
cented raisiin, honor or honur from 
honor-em honur, etc., with a very 
forcible impact of the voice upon the 
last syllable. 5 The effect of this pre- 
dominant influence of the accented syl- 
lable would necessarily be, the trans- 
formation of the atonic syllables.^ We 
see evidence of this result in the not 
unfrequent appearance of lienor, enor, 
and annor in the place of honor honkr. 
An instance, however, perhaps bearing 
more directly on our present purpose, 
is afforded by the derivatives of the old 
French or Norman coer or cuer (coeur). 
There is little doubt that this was 
originally pronounced (kuur).'' When, 
however, by the addition of -age, there 
resulted cordge, curdge, and courdge, all 
13. forms, both the quantity and qua- 
lity of the original (uu) was affected, 
and almost of necessity the atonic 
cor, cur, cour, would become (kw), and 
the entire word (k^raadzh-^). In the 
process of development cordge next 
receives the syllable -os or -us, and 
becomes coragos, coragus curagos, or 
curagiis, all of which are admissible 
Norman forms. The lately long vowel 
a is now changed both in quantity and 
quality, and has become (e, i, o) or 
(a, ■b), it is not easy to say which, and 
the result may be probably considered 
as (kareguus*).^ Similarly it might be 
shewn that curt cour = (kuurt), becomes 

the old Norman system of accentuation, and 
that evidently adopted by Chaucer, which 
agrees with classical French, supra p. 331. 

6 Admitting that this obscuration of un- 
accented vowels often occurs, and has been 
especially active in many languages, I must 
deny it to be a necessity of pronunciation, 
any more than the prolongation of a vowel 
by the accent, witness the clear unaccented 
but extremely short a, and the decidedly 
short but accented o in the Italian am5 
(amo-). See infra p. 585, n. 4. — A.J.E. 

■^ Not having sufficiently studied Norman 
orthography and pronunciation I am un- 
able to speak on this point. — A.J.E. 

8 It seems to me extremely doubtful that 
such a sound as (a) was known to the Nor- 
mans, when regard is had to its very late 
introduction into England, suprst p. 172. 

Chap. VI. § 1. 



eurteis (ktjrtees*), and this again ewr- 
teisie (kortesii-t?), or perhaps, at least 
occasionally, (kartesee-).^ The last 
word became, as is well known, in 
English curtesie, cortaysie, courtaysie, 
all of them, by the above theory, being 
pronounced (kortesii-'c) or (kBrtesee'e), 
or very nearly, accentu mutato, as ths 
modern courtesy, that is (kor'tesi).^ 
The spelling could not on this theory 
have affected the pronunciation,-^ which 
was determined by the power of the 
tonic accent obscuring and transform- 
ing the independent value of the atonic 
syllables. It may further be observed, 
that the u in the former cur, being so 
close to the predominant accent, be- 
came positively eclipsed by it, and 
would therefore be exceedingly short 
and obscure, as (b) in English, while 
the u in the second cur, receiving a 
secondary accent, would probably have 
a clear and definite sound, equal to 
(kar). It is this sound which the 
English derivatives would receive when 
no longer under the influence of the 
Norman accentuation, but subjected to 
the entirely different system of the 
English. Hence the Norman : jurnee, 
trubler, colur, cumfort, suverain, doz- 
aine, covert, custume, dobler, curtine, 
hurter, cumpainee, turnoiement, sujur- 
ner, sucur, etc., when they became re- 
spectively : j6urney, trouble, colour, com- 
fort, sovereign, dozen, covert, custom, 
double, curtain, hurt, company, tourna- 
ment, sojourn, succour, etc. would 
naturally be pronounced very nearly as 
they now are, or very recently were.* 
In the present sound then of these 

words, we see the Norman influence 
still persisting.^ Exceptions may no 
doubt be taken to this general assertion, 
but the main principle can hardly be 
affected by them. It may be further 
remarked, that the continual inter- 
change in early English, of u, e, i, in 
such instances as : werk wirk, chirche 
cherche churche, kirtel kertel kurtle, 
erth urthe, sunne sinne, sturn stern, 
cherl churl, segge sigge sugge=sai/, in 
bdt/md, etc., compared with bathed, 
etc., in tellies for telles, ledus and ledys 
for ledes, and in such plurals as femdlus, 
sydus, coupus, (see Anturs of Arther 
passim,) tends to shew that the short u 
had the same sound both in Norman 
and English.^ It is impossible to con- 
ceive that tbe unaccented us, which 
merely stands in these instances for -es, 
was pronounced (us). It must have 
had the same obscure sound as the u in 
eurteis. When, however, this obscure 
unemphatic sound is required to take 
the accent, then it assumes the clear 
utterance of the u in curtesie. Hence 
the u in churche, urthe, sunne, sugge, 
was not unfrequcntly found inter- 
changing with e and i short. The 
sound then of short u seems, in 
words of more than one syllable, to 
depend on the principal accent, and 
when atonic to be (9), and this was 
also the sound in monosyllables na- 
turally short, as church, churl, etc. The 
merits of the general theory, which I 
have here attempted to expound, can, 
however, hardly be fairly judged of by 
this brief and imperfect representation 
of it. 

I do not feel satisfied that the ahove ac- 
count of the successive formations of coeiir, 
courage, courageux, is historically correct. 
— A.J.E. 

1 If this termination were ever (-ee), it 
■was only through the West Midland con- 
fusion of i, e, and rejection of final e, cer- 
tainly not from reading ie as ei, and calling 
that (ee). It was dialectic, not literary. — 

« The absolute ignorance of the sound (a) 
shown by all the authorities of 16., makes 
me inclined to reject at once the hypothesis 
that courtesy could have been called (kar*- 
tesi) in 14. With regard to the second syl- 
lable of the word, more is said in Chap. 
"VII. \ 1, near the beginning.— A.J.E. 

* Although after the invention of print- 
ing, spelling may have affected pronuncia- 
tion, in 12. 13. and 14. we have no reason 
to assume anything but the converse, 

namely, that pronunciation affected spell- 
ing.— A.J.E. 

* But they were not so pronounced in 16., 
as we know by direct evidence, and they 
are not now so pronounced by tbe illiterate 
in our provinces. It was only the other 
day that I heard a porter at Clapham 
Junction shouting out many times in suc- 
cession (Klap'am Dzhz<q'shwn). with pure 
(m) and not (a), and without any obscura- 
tion of the unaccented vowels. — A.J.E. 

s The history of the introduction of (a) 
being now on record, and the battle be- 
tween (a, u) being still undecided, I do not 
see how this conclusion can be admitted. 

6 See supra p. 299, and 300, n. 2, also p. 
425, p. 507 and numerous instances in Chap. 
V. \ 1, No. 3. But there seems no reason 
for supposing this u to have been anything 
but (y, e, i).— A.J.E. 


MR. PAYNE ON OE, 01, UE, UI. Chap. VI. § 1. 

OE, EO (in one Syllable), 01, 10 (in one Syllable), UE, EU, UI, lU 
(in one Syllable), and the variants EOU, EOW, EOUW, EW, 
lEU, IW, JEW, W, UW, EACH = (uu). 

The illustrations and arguments by 
which the above proposition is sup- 
ported, are given at some length in my 
paper. A brief summary, which under- 
states the proof, is all that can be given 
here. Assuming that Norman long or 
tonic M = (uu), it was ascertained ^ that 
Norman ui, and inferred ^ that the in- 
verted tKf had the same sound as u 
alone, that is, that miit = (jiuut), fruit 
= (fruut) ,^ riule = (ruul-e) . These con- 
clusions depend on the light shed by 
Norman and English on each other.* 
Thus in English texts frute rhymes 
with dedute, i.e. Norman deduit, and 
again frui with dedwt, whence ui 
= u = w = (uu). Again Norman 
suir, siur to follow, becomes siw in 
Layamon, sinve in Ancren Eiwle, 
swe in the Alliterative Poems, and 
sewe in Chaucer, shewing ew, ui, iu, 
tw, uw = {mi), and therefore sewe of 
Chaucer = (suu-e). 5 The argument thus 
gained, applied to triw-e (Robert of 
Gloucester), trewe (Chaucer), truwe 
(Occleve), and treiie (Audley), gives 
theoretic (truu-ej, which is shewn to 
be correct by trtve in Alliterative Poems, 
p. 27, where due also rh}Tnes with it, 
supported by Promptorium Parvulorum 
irweS" Thus, in addition to the digraph 
above given, ue and eu also appear to 
= (uu). If then the ags. treoive, which 
appears as treowe and treouwe in Laya- 
mon' s earliest text, and as trewe in the 
later, had a sound different from trewe, 

1 The proof must be sought in the paper 
referred to, and having not seen it, I can 
only express my own doubts of its correct- 
ness founded upon my own small amount 
of obserA'ation, see p. 458. — A.J.E 

2 Apparently from the theory that an 
inversion of the order of the letters in a 
digraph does not affect its value, which is 
to me extremely doubtful. — A.J.E. 

8 In nuit, fruit, the i, still pronounced, 
is as much a representative of the lost gut- 
tural, as the y in day, may. — A.J.E. 

* Which I doubt.— A.J.E. 

6 An examination of the age and locality 
of MSS. is necessary before judging of the 
value of their orthography in determining 
sounds. The Alliterative Poems, Sir 
Gawayne, and Anturs of Arther are West 
Midland, in which part of the country a 
very peculiar pronunciation still prevails, 
so different from the South Eastern, that 
the ancient orthography of that district re- 
quires especial study. It is very probable 
that (uu) was unknown in those districts as 
a sound of u, w, but that it was always 
replaced by (yy, y) or some cognate sound. 

triwe, or trwe, it could only have been 
for a short time, and it may probably 
be assumed to have been the same.'' 
The supposition, then, that ew had one 
sound in words of Norman origin, and 
another in those of native growth (p. 
302) is unnecessary, and indeed incon- 
sistent with the fact that, though it 
may be true that Chaucer does not 
rhyme together words in ew of different 
origin, other writers do. As a case in 
point we find in Alliterative Poems, p. 
13, trwe English, blwe probably Nor- 
man, greive preterit, remwe Norman, 
and again knewe English, (which is 
also found written knwe) swe Norman 
dne Norman, hwe English, untrwe Eng- 
lish and remwe Norman, all rhyming 
together.^ "We note also in this text 
Chaucer's neive always spelled nw or 
mve. We should, therefore, perhaps 
read such rhymes as those found 
in Lyrical Poetry, p. 37, viz : reowe, 
newe, heowe, kneowe ; as (ruu*e, nuu"e, 
Huu-e, knuu-e). Many confirmatory 
instances might be cited from various 
texts, but the above may suffice to^shew 
the great probability that Norman and 
English ue eu, ui iu, eou, etc. were in 
13. and 14. = (uu), and hence that the 
modern pronunciations of : rue, true, 
sue, suit, rule, pursuit, bruit, fruit, and 
the vulgar sound of: nuisance (nun), 
duty (duu), new (nun), beautiful (buu), 
are but echoes of that of 13. and 14.^ 

On Layamon see p. 496, and on the Ancren 
Riwle, seep. 506. The orthography of these 
works offers so many points of difhculty 
that it cannot be safely appealed to for any 
proofs. The whole of our Western provin- 
cial pronunciation has first to be studied.— 

6 In the last note it was conjectured that 
the IV of the Alliterative Poems may have 
been (yy). As regards the Promptorium 
the author only knew the East Anglian pro- 
nunciation (supra p. 23, note 2), and to this 
day the East Anglians use (yy) for (uu). 
The above inference is therefore in the 
highest degree hazardous. — A.J.E. 

7 On treoive see p. 498, 1. 14. No Anglo- 
sax on scholar would be likely to admit eo 
to have had the same value as u. See p. 
511.— A.J.E. 

8 Probably all these rhymed as (yy), as 
they still would in Devonshire. See* supri 
n. 5.— A.J.E. 

9 This conclusion is directly opposed to 
all I have been able to learn on the subject. 
— A.J.E. 

Chap. VI. § 1. MR. PAYNE ON 01, 10, OE, EO. 


01, 10 (in one syllable), OE, EO = (uu) or (ee). 

It is remarkable that two sounds so 
remotely allied as (uu) and (ee) should 
frequently, both in Norman and Eng- 
lish, be used one for the other. No- 
thing, however, is more probable than 
that oi in early French generally, 
must have represented the sound (uu). 
Nothing at the same time is clearer 
than that in the Norman texts the 
oi of Central France is very gene- 
rally to be read (ee). Thus the forms 
moij toi, etc., which in proper Nor- 
man would be mei, lei, etc., are by 
no means excluded from Norman texts, 
but are constantly found rhyming with 
the Norman ei or ee. Thus tei rhymes 
with moi, moi with/o«, voir with veer, 
rot with lei, etc., and are therefore to 
be pronounced (mee, veer, lee), etc. 
The concurrence, however, of such 
forms as : genoil genou, genoul, genue ; 
acoiller, acuiller, where ui = (uu) ; agoille 
aiguille; angoisse, anguisse, angusse ; 
noit, nuit ; poi,pou peu ; fusoxjn (rhym- 
ing with corhiloun in De Biblesworth, 
"Wright p. 158), seems to shew that oi, 
w = (uu). This conjecture may be 
further confirmed by assuming o{ = o^, 
and observing that oile oil of 12. be- 
comes oele and xiille in 13., and huile 
in 15., while buef, boef are bouf= 
(buuf) in De Biblesworth. This word 
he rhymes with ouf ceuf, of which the 
variants were oef, uef. Again boe, 
moe, roe of 13., become later boue, moue, 
roue. But eo also = (uu), as is seen in 
the numerous words of the form etn- 
pereor, etc., which became emperour, 
etc. The most difficult case is that of 
io = oi={\m). It is proved, however, 
by the formation of such words as 
mansion, which became by the loss of 
the n and fusion of io into u, maisun. 
Raisun may be explained in the same 
way, as may also magun mason, from 
low Latin macio. The word in its 
Normanised form machun occurs in 
Layamon, and is erroneously translated 
machine by Sir F. Madden. These 
views respecting Norman oi io, oe eo = 
(uu), are singularly confirmed by Eng- 
lish examples of adopted Norman words. 
Mr. Ellis's inferences (p. 269) I should 
generally endorse, except that, as before 

1 In this further investigation respect 
would have to be paid to the principle of 
palatalization produced by an inserted i, 
familiar to those who have studied phonetic 
laws, and well illustrated by Prof. H aide- 

stated, I should pronounce boiste, for 
which buiste is also found (buust'e) not 
(buist'e), and perhaps Loi, cog, and bot/ 
(Luu, kuu, buu). Merour mirror of 
Chaucer, is directly taken from Norman 
mireor. It occurs as myroure in Po- 
litical Songs, Wright, p. 213. Norman 
poeste also appears constantly in English 
as pouste. The case of io = (uu) is not 
considered by Mr. Ellis. It is, how- 
ever, rendered more than probable by 
our word warrior written werroure by 
Capgrave, and referable to Norman 
guerreur, which by analogy =guerrour. 
Analysing the o« = (uu) into oi= io, we 
obtain the modern English warrior. 
Similarly we may trace carrion to Nor- 
man earoine. So the word riot, con- 
jecturally referred by the editor of 
Ancren Riwle to route, may be really a 
variant of that, word. it must be 
remembered, however, that the English 
riot came directly from Norman riote, 
and the variation, if variation it be, 
must have belonged to the original 
source. Diez, Menage, Scheler and 
Burguy virtually give up the ety- 
mology altogether- It is only probable 
then, but not proved, that Norman 
earoine and English carrion, might 
have been (k«ruune), and that riot 
might have been sometimes (ruut). 
The subject requires further investi- 
gation. ^ The fluctuations of Norman 
orthography suggested the enquiry that 
has been sketched, but the results lead 
us on still further, and render it pro- 
bable that eo, oe, etc., when found in 
pure English words, had also the 
sound (uu). Heo she, therefore, with 
the variants hu and hue, was probably 
(huu), as it still is in Lancashire. 
Meore their, too, and huere, interpret 
each other, and so do, duere and deor, 
beoth and bueth, beon and buen, preost 
and pruest, glew and gleo. AVe infer, 
then, that in Layamon' s beorn warrior, 
cheose, leode, leof, leose the eo — {m\). 
The subsequent forms burn (Piers Plow- 
man), choose, luve, loose, etc., and the 
contemporary form lued for leod, (Pol. 
Songs, p. 155), render this hypothesis 
very strong, while such forms as goed 
good, compared with goud (Layamon,) 

man, in his article on Glottosis Analytic 
Orthography, pp. 67-71. So far as I can 
understand them, I entirely dissent from 
the views expressed in the text.— A.J.E. 



Chap. VI. § 2. 

toen town, proeve Norman preove Eng- 
lish, doel and r/to/ sorrow, shew that oe as 
well as eo = (lui). The great difficulty in 
assigning the phonetic values of oi, eo, 
oe arises from the undoubted fact that 
they were r('pr(!sented both by (uu) and 
by (ee).^ Thus Ave find that nearly 
all the Norman and English words 
cited above appear to have both sounds. 
Thus Jieo appears as he, heore huere as 
here, deor dnere as dere, heoth bueih as 
beth, heon biien as ben, preost pruest as 
prest, chcose as chese, leose as lese, etc.,^ 
also proeve preove as preve^ caroine as 
careyne carayne, piiple, pueple, people 
as peple, etc. This divarication in 
the case of Norman words, was more 
apparent than real, since the usual 
Norman sound of oi was (ee). Yet 

the numerous examples of oi al80 = 
(uu), as for instance in the normal 
termination of the third person sin- 
gular of the imperfect tense of the 
first conjugation, which was -out = (uut), 
while in the other conjugations it was 
-eit = (-eet), render the determination of 
the law of divergence very difficult. 
This law, however, must apparently 
have equally dictated the interchange 
of the sounds as well in English as in 
Norman, and this fact is only one proof 
more of the remarkable correspondence 
(in spite of all orthographic variations) 
between the phonetic systems of the 
two languages, and illustrates the ge- 
neral position that the Norman and 
English pronunciations respectively 
help to determine each other.* 

§ 2. The Expression of the Sounds. 

The list in tlie last section suggests its counterpart, how 
have the sounds of the English language been expressed by 
letters at different times ? Up till the invention of printing 
at least, the object of writers seems to have been to represent 
their pronunciation, and the possibility of using the same 
symbols with altered values does not appear to have occurred 
to them, although each sound was not uniformly represented 
by the same sign, and some signs had more than one value." 
It is also not at all improbable that very provincial writers 
may have been accustomed to attach values to the letters 
corresponding to their local pronunciations, and have then 
used them consistently according to their lights. From 
these causes arose the occasional picturesqueness of scribal 
orthography, which was unchecked by any acknowledged 


1 My own indicated explanation of the 
phenomena to which Mr. Payne refers are 
to be found on p. 2G9, and 131, note, col. 1, 
p. 138, note col. 1. The question seems to 
be one affecting the treatment of Latin e, o, 
in the Romance languages. — A.J-E. 

2 These anomalies, occurring in MSS. not 
expressly named, seem readily explicable 
by the known interchanges of eo, e, p. 488, 
and of u, e, supra p. 585, n. 6.— A.J.E. 

8 Oe, eo are so rare in Chaucer, see p. 262, 
1. 33, tiiat I have not been able to judge of 
their origin or intentional use as distinct 
from (ee). But we must not forget the two 
modern forms reprove, reprieve. — A.J.E. 

* The Norman was an old Norse phonetic 
system modifying the langue d'oil, so that 
the latter had the main share in the result. 
The English was a pure Anglosaxon system, 
slightly modified by an old Norse element. 

There seems to be no connection between 
the two systems of sound. The orthogra- 
phies were both derived from the Latin, 
but the Norman spelling came direct from 
Roman sources, and the Anglosaxon was 
only a priestly transcription of the pre- 
existent runic. The whole application of 
the orthographies was therefoi'e diverse. 
The Norman accidentally came into collision 
with the English, but the developments 
seem to have proceeded independently, and 
the share of Norman in 13. English was 
scarcely more than that of English in 13. 
Norman. Ultimately the whole character 
of our language, both in idiom and sound, be- 
came English, and Norman words were ruth- 
lessly anglicised. Hence, I am not inclined 
to admit Mr. Payne's conclusion. — A.J.E. 

5 See the table on p. 407, where in col. 
2, " (ou) 00 oa'* is a misprint for " (oo) 
00 oa." 


authority. At the present day we have nothing to guide 
us but the usage of printing offices, on which (and not 
on the manuscripts of authors) our orthographical laws and 
the pages of our dictionaries are founded. The most in- 
geniously contradictory reasons are given for preferring one 
spelling to another. Sometimes a man with a name, as 
Johnson in England and Webster in America, proclaims his 
own views and is considerably followed, but Johnson's favourite 
'ick as in miisick has disappeared, and no Englishman likes 
to see the American orthography.^ Daring the last fifty 
years a habit of eye has been generated, and spelling has 
been dissociated in our minds from the expression of sound. 
But even in the xvth century this was not the case in 
England, although the disappearance of final e from pro- 
nunciation introduced more and more confusion as the 
century advanced, and the original value of the e was less un- 
derstood. When printing commenced, there was a necessity 
for printers to introduce some degree of uniformity, and, as I 
have had personal experience of the difficulties thus created,^ 
I can well understand the slowness with which even toler- 
able uniformity was attained. It took fully two, if not 
three, centuries to reach the present system. During this 
time several experiments were made, among which I do not 
reckon schemes for an entire renovation of our orthography, 
as proposed by Smith, Hart, Bullokar, Gill, and Butler, in 
the first century and a half after Caxton set up his press. 
The last great change was made in the xvi th century, when 
the orthographies ee ea, oo oa, were settled (pp. 77, 96), 
how, and by whom, I have not yet discovered. The intro- 
duction of ie, in place of ee, was not of the same nature, and 
did not take root till the xviith century (p. 104). In the 
course of that century many little changes were tried, but 
the gradual loss of the feeling for the meaning of ea, and its 
perversion in the early part of the xviti th century (p. 88), 
undid most of the good effected in the xvi th century. No 

^ Since the publication of the Die- suivies d'une histoire de la reforme or- 
tionary of the French Academy, it has thographique depuis le XV® siecle 
become the sole rule in France, or jusqu' a nos jours, 2nd ed. Paris, 1868, 
rather each of its six editions of 1694, 8vo. pp. 485. 
1718, 1740, 1762, 1795, 1835, has be- 
come the rule till certain points were ^ jn 1848-9 I conducted a phonetic 
reconsidered and changed in subsequent printing office with a view of trying 
editions. '' Le Dictionnaire de I'Aca- the experiment of a phonetic ortho- 
demie est done la seule loi," says the graphy, and I had to drill compositors 
most competent authority in France, of all kinds of pronunciation to a uni- 
M. Ambroise P'irmin Didot, in his ex- form system of spelling, in order that 
tremely interesting Observations sur all my books, and all parts of my 
rOrthographe ou Ortografie fran(,'aise, books, should be consistent. 



Chap. VI. § 2. 

great change was effected by Johnson over Dyche and 
Buchanan, but he became a name, and a refuge for the 
printer's reader. We have not yet settled how to write 
between two and three thousand of the words in our lan- 
guage,^ although it must be confessed that we do not find 

new edition of the dictionary of the 
Academy, says : "L'usage si frequent 
quej'aidu faire, et que j'ai vu faire 
sous nies yeux, dans ma lonj^ue carriere 
t}^ographique, du Dictionnaire de 1' 
Academic, m'a permis d' apprecier 
quels sont les points qui peuvent oflfrir 
le plus de difficultes. J'ai cru de mon 
devoir de les signaler. L'Academie 
rendrait done un grand service, aussi 
bien au public lettre qu' a la multitude 
et aux etrangers, en continuant en 1868 
Tceuvre si hardiment commencee par 
elle en 1740, et qu'elle a poursui\ie 
en 1762 et en 1835. II suffirait, 
d'apres le meme systeme et dans les 
proportions que 1' Academic jugera con- 
venables : 1° De regulariser I'ortho- 
graphe etymologique de la lettre x? c^ ; 
et de substituer aux 0, th, et <^, ph^ nos 
lettres fran^aises dans les mots les plus 
usuels; d'oter 1' h a quelques mots oil 
il est reste pour figurer 1' esprit rude ('); 
2" De supprimer, conformement a ses 
precedents, quelques lettres doubles qui 
ne se prononcent pas ; 3° De simpli- 
fier I'orthographe des noms composes, 
en les reunissant le plus possible en un 
seul mot ; 4"^ De regulariser la desi- 
nence orthographique des mots ter- 
mines en ant et ent ; 5° De distinguer, 
par une legere modification (la cedille 
placee sous le t), des mots termines en 
tie et tion, qui se prononcent tantot 
avec le son du t et tantot avec le son 
de I's ; 6° De remplacer, dans certains 
mots, Vy par 1' i ; 7° De donner une 
application speciale aux deux formes 
^ et ^ au cas ou le j, dont le son est 
celui du g doux, ne serait pas preferable ; 
8° De substituer Vs a Vx, comma 
marque du pluriel a certains mots, 
comme elle I'a fait pour lois, au lieu de 
loix [lex, la loi, legefs, les lois). Parmi 
ces principales modifications generale- 
ment reclamees, 1' Academic adoptera 
celles qu'elle jugera le plus importantes 
et le plus opportunes. Quant a celles 
qu'elle croira devoir ajourner, il suj9fi- 
rait, ainsi qu'elle T'U fait quelquefois 
dans la sixieme edition, et conforme- 
ment a I'avis de ses Cahiers de 1694, 
d'ouvrir la voie a leur adoption future 

^ E. Jones, The common sense of 
English Orthography, a guide to the 
Spelling of doubtful and difficult words, 
for the use of printers, authors, ex- 
aminers, teachers, and students gene- 
rally, 1867. It may be observed that 
he puts printers first. He lays down 
as "the principles of English ortho- 
graphy," first, "the law of abbrevia- 
/ tion or contraction," illustrated by 
music, blest, things, inferior, baking, 
entrance, xcilful, fetter, for musick, 
blessed, thynges, inferiour, bakeing, en- 
terance (?), willfull, feetter (?), second 
" preference for, or aversion to, certain 
letters illustrated by the disuse of y in 
middle, and use of it at the end of 
words." The statement that "the 
desire to produce an agreeable succes- 
sion of sounds, or euphony, is also an 
important principle in the spelling of 
words," is unintelligible in an ortho- 
graphy which does not regulate the 
sound. He classifies the doubtful 
words thus : ] . honor, honour (30 
words) ; 2. movable, moveable (Johnson 
inconsistent) ; 3. civilise, civilize ; 4. 
traveler, traveling, traveled, traveller, 
travelling, travelled ; 5. enrol, enroll ; 
6. pressed, dressed, prest, drest; 7. 
medifeval, medieval ; 8. monies, mo- 
neys ; 9. hinderance, hindrance; 10. 
alcali, alkali ; 1 1 . Frederic, Frederick ; 
12. connection, connexion; a license, 
to license, advice, advise ; 14. centre, 
center ; 15. bark, barque ; 16, tong in 
XVI th century, tongue; 17. controul, 
control. And he then proceeds to give 
rules for spelling in these doubtful 
cases. His arguments do not merely 
aflfect the words he cites, but large 
numbers of others which he does not 
presume to alter, because they are not 
considered doubtful. This is the most 
recent attempt at giving "principles" 
to regulate our orthography. The 
reader will find a Report on this work 
by Mr. Russell Martineau, in the 
Transactions of the Philological Society 
for 1867, Part IL, pp. 315-325. M. 
Didot, in the work cited on p. 589, n. 
2, in anticipation of a revision of 
French orthography in a contemplated 

Chap. VI. § 2. 



much inconvenience from the uncertainty, and most writers 
select the spelling which their hand takes from habit with- 
out consideration, and do not call the compositor to order 
if he alters it in print. And compositors, with their 
authorized superiors, the printers' readers, have habits of 
their own as to spelling and punctuation, regarding their 
author's MS. as an orthographical exercise which it is their 
business to correct ; so that, except in extremely rare cases 
where the author is opinionated and insists on the com- 
positor " following copy,'' ^ no printed book represents the 
orthography and punctuation of the man of education 
who writes, but only of the man of routine who prints.^ 

au moyen de la formule : Quelques-wts 
ecrivent . . , : ou en se servant de 
cette autre locution : On pourrait 
ecrire .... Par cette simple in- 
dication, chacun ne se croirait pas irre- 
vocablement enchaine, et pourrait ten- 
ter quelques modifications dans I'ecri- 
ture et dans r impression des livres," p. 
23. This is the latest French view of 
the question. 

1 And then the compositor can 
easily take his revenge, and disgust his 
author, by copying all the careless 
blunders which haste and the habit of 
leaving such matters to the printer 
have engendered in our writers. The 
literal exhibition of the greater part of 
*' the copy for press," and still more of 
the correspondence, of even esteemed 
men of letters, would show that our 
present orthography, including the use 
of capitals and punctuation, is by no 
means so settled as printed books, and 
the stress laid upon " correct " spelling 
in Civil Service Examinations, would 
lead us to suppose. 

2 Some months after this paragraph 
was written, I received a letter from 
Prof. F. J. Child, of Harvard, in 
which he says : " I wish you may make 
the Philological Society take some 
tenable ground as to orthography in 
their dictionary. Nothing can be 

jmore absurd than the veneration felt 
and paid to the actual spelling of Eng- 

llish, as if it had been shaped by the 
national mind, and were not really im- 
posed upon us by the foremen of some 
printing offices. In America all books 
printed in New York exhibit Webster's 
spelling, and most books printed at 
Cambridge (a great place for printers), 
"Worcester's. Although we cannot 
trace the English spelling-book, so far 

as I know, I am fully convinced that 
it is largely of printing office origin." 
As this sheet was passing through 
the press my attention was directed to 
the following letter from the Mr. Jones, 
mentioned p. 590, n. 1, in the Athe- 
naeum, 10 July 1869, in which he 
seems to be endeavouring to give effect 
to his views by means of an association. 
The "Fonetic Nuz " Spelling alluded 
to, is that employed by the present 
writer in the Fhonetic News in 1849 : 
'■^Spelling Rej'ornt. — Perhaps you will 
allow me a short space to lay before 
your readers a brief statement of the 
objects of the Spelling Eeform Asso- 
ciation. The very mention of ' Spelling 
Reform' suggests to most people some- 
thing like the ' Fonetic Nuz ' system, 
which has been the subject of so much 
ridicule. Permit me then to say, with- 
out expressing any opinion upon the 
phonetic method, that the Spelling 
Reform Association does not propose to 
introduce that mode of Spelling the 
English language, but that our recom- 
mendations are based upon the follow- 
ing assumptions, which most persons 
will readily admit: — 1. No one would 
desire to stereotype and hand down to 
posterity our orthography in its pre- 
sent state ; but there is a vague 
notion that at some time and by some 
means the thing will be rectified. 2. 
England is about the only country in 
Europe in which the orthography has 
not been, in some way or other, ad- 
justed ; and orthography is one of the 
very few subjects in England which 
have not been adapted to modern re- 
quirements. 3. The anomalies of the 
orthography cause serious obstruction 
to the education of the people, most 
of the time in Government schools 



Chap. VI. ^ 2. 

Still there is a latent spark of that fire which warmed the 
original writers of our own manuscripts, and there is a 
notion that certain combinations have an inherent tendency to 
represent certain sounds, and conversely that certain sounds 
are naturally represented by certain combinations. The last 
section will have shown with what allowances the first state- 
ment must be received in the xix th century ; the following 
table will show how varied are the combinations which have 
been and are employed to represent the sounds. 

In drawing up the list of sounds represented, it was 
necessary to include all the sounds which, so far as the 
preceding investigation shews, previously existed in our 
language, and those which recent and minute examination 
establishes to exist at present, including those newly in- 
troduced French words which are spoken in a semi-French 
pronunciation. The following list is an extract from the 
completer list of spoken sounds in the introduction, and for 
convenience is arranged in the same order. The same 
abbreviations are used as in the last section. 

being occupied in teaching reading and 
spelling — with arithmetic — with miser- 
able results, as to the proportion of 
children turned out of these schools 
having the ability to read with intelli- 
gence and to spell correctly. 4. The 
various examinations conducted by the 
Government, the Universities, and other 
examining bodies, give a fictitious value, 
and virtually give the sanction of their 
approval, to a system which has no 
claim Avhatever to be regarded as * the 
best method of spelling words,' a sys- 
tem w^hich has been described by high 
authority as ' an accidental custom, a 
mass of anomalies, the growth of ig- 
norance and chance, equally repugnant 
to good taste and to common sense.' 
5. A simplitication of the orthography 
would do more to give the people the 
ability to read with intelligence and to 
spell correctly than any amount of 
Government grants or any legislation 
whatever. 6. No individual or society 
under present circumstances would 
have sufficient influence to introduce an 
improved system of orthography ; if 
done at all, it must be by the co-opera- 
tion of literary men. teachers, examin- 
ers, printers, and the public generally. 
7. It is possible, by observing analogy 
and following precedent, wathout in- 
troducing any new letters or applying 
any new principle, to simplify the or- 

thography so as to reduce the difficul- 
ties to a minimun, and to replace con- 
fusion and caprice by order and symme- 
try. The Spelling Reform Association 
invite the co-operation of all literary 
men and friends of education in this 
desirable object. E. Jones, Hon. Sec." 
The opinions entertained by the pre- 
sent writer on the subject thus broached 
by Prof. Child, and Mr. Jones, will 
be developed in the subsequent sections 
of this chapter, and the same remarks 
apply mutatis mutandis to M. Didot's 
French proposals. It will there ap- 
pear that I do not see how any " tenable 
ground" can be taken by the Philo- 
logical Society "as to the orthography 
of their dictionary," beyond the accident 
of present custom in London. Much 
might be said on Mr. Jones's seven 
points, which he believes "most per- 
sons will readily admit." "Why our 
present orthography should be con- 
sidered so much less worthy to be 
handed down to posterity than one 
modified on Mr. Jones's "principles," 
and how any such modifica.ions would 
render its use beneficial in schools to 
the extent anticipated, I am at a loss 
to conceive. To Mr. Jones's seventh 
proposition, if I understand it aright, 
my own orthographic studies lead me 
to give an unqualified denial. 

Chap. VI. § 2. 



Chronological Account of the Expression or Sounds in English 


(A a), was always represented by a 
from 13. to 19., the sound went out 
in 17., and now only exists in rather 
a rare pronunciation of: «sk, st«ff, 
commatnd, p«ss, and similar words, 
and is considered to exist in : star, 
card, by those who believe the vowel 
short ; it is common in the provinces 
in place (se). 

[A a), was probably the ags. sound 
written a, possibly the sound meant 
by oa in 13. ; it is now lost in Eng- 
lish, but is heard in Scotch. 

(:A a), according to Wallis, etc., the 
sound into which short o fell in 17. 
when " fall folly, call collar, lawes 
losse, cause cost, aw'd odd, saw'd 
sod," were considered as perfect 
pairs. In 19. this short o is (o). 
The distinction is delicate, but may 
be rendered appreciable by drawling 
odd into (ood) which will be found to 
be diiferent from awed (AAd), or by 
shortening the vowel in the latter 
word, producing (Ad) which is dif- 
ferent from odd (od). In 19. a after 
a (w) sound, as ivhat^ watch, squash, 
(whAt, wAtch, skwAsh), is the sole 
representative of this sound, and 
even here most speakers use (o). 

(Aa aa) was represented by a always 
in 13., and by a in open, and fre- 
quently by aa in closed syllables in 
14. In 16. it was still written a 
without any indication that the syl- 
lable was long, except by an occa- 
sional mute final e. The sound was 
lost in 17., except perhaps before r, 
so that ar, er in iar, cl^rk, may have 
represented (aar), though they were 
acknowledged, and perhaps most fre- 
quently pronounced, as (a3r) only. 
In 19. the indication of length and 
quality is variously made according 
to the origin of the word m : father, 
axe (but not in bare, fare, etc.), 
sera^^lio, ah, a/ms, Ma^mesbury, ecla^, 
aunt, b'/rq2«e, ch;rk, heart, gward, but 
its principal indication is a before 
r— [x) professedly, but intended to be 
omitted by those persons who write 
larf to indicate (laaf). In London 
ar, when not followed by a vowel, 
may be regarded as the regular sign 
for (aa), and is so used by many 
writers. The ah ! of the exclama- 
tion is, however, nearly as certain, 
and does not involve the r difficulty. 

{Aa aa), this appears to have been the 
long a of ags. It has since disap- 
peared from acknowledged sounds. 
It is, no doubt, heard in the pro- 
vinces, and it is by some recognized 
as the common London sound meant 
for (aa), which see. 

(:Aa aa), unknown previously to 17., 
and then represented by au, aw, 
augh, ough ; these sounds and nota- 
tions still prevail. It replaced the 
sound of (au), and hence was repre- 
sented by a before I, as now ; or by 
al, with a mute I. It was identified 
with the German a, and is often 
called " German « " in pronouncing 
dictionaries ; it was also identified 
with French a, and Miege could 
not hear the diff'erence. See Eron- 
dell's remarks supra p. 226, n., col. 2. 
In 17. oa represented it in broad. The 
following may be considered as its 
representatives in 19. : fall, aara, 
Magdalen College (MAAd-len), maul- 
stick, wa^k, bazJman, hawl, Ma^^de, 
naiighiY, Yaughan, auln, awful, 
awe, broad, so/der (spelled sa^^der 
in Sam Slick), ought. The com- 
bination or is theoretically (oi), 
practically (aaj), or (a a) ; so that 
Dickens, in Pickwick, writes Smorl 
Tork as a name to indicate synall 
talk. See supra, p. 575, under o. 
Hence, extraordinary, Georgia, 
George, fork, horse, may be reckoned 
as other examples, even by those 
who do not include the r in the 

(Aah aah). This delicate sound pro- 
bably formed the transition from 
(aa) to (sese) in 17., and it is occa- 
sionally heard from "refined" 
speakers, as a variety of (aa), which 
they consider too "broad," while 
(seae) used by others is too " minc- 
ing." It is a mere variety of (aa), 
and is represented in the same way. 

{M se) was probably the short ags. ae, 
but in ags. it rapidly became con- 
fused with (e, e), and was then 
lost. It reappears in 17. as a sub- 
stitute for (a), and was represented 
by a and the same varieties as that 
sound. So it has remained, but by 
omitting letters, and reducing many 
(aa), and even other sounds, to this 
favourite short vowel, it is seen va- 
riously represented in 19., as: sat, 



EXPRESSION OF SOUNDS. (iEsD— D). Chap. VI. $ 2. 

Isaac, Mackay, drachm, have, always 
(naav) down to 16., brt^nio (ban-Jo), 
Tag/imon (Tyom-Bn), pl««d, salmon, 
harangue, Clap/<am, considered as 
(Kla'p-iem), but really (Kluep'Bm), 
Tolkmache (Taol-msesh), piqwant. 
In 17. one, once were (waen, wa^ns). 
It is in 19. also used by very delicate 
speakers, especially educated ladies 
in Yorkshire, in such words as : 
basket, staff, path, pass, awnt, in 
which (ah, a) and (aeae, aah, aa) are 
also heard. This vowel is now cha- 
racteristic of English, and is the 
despair of foreignei-s. 

(^33 a38e). The long (aeae) replaced 
(aa) in 17 , and was represented in 
all the ways in which (aa) had been 
previously pronounced. No change 
was acknowledged. The sound ra- 
pidly died out into the (ee) of 18., 
but it is now preserved in the "West 
of England, where (Bajajth, ksesead) 
are pronounced for Bath, card. It 
is the name of the letter A in Ire- 
land. Twenty years ago it was, and 
probably still is, a fashionable long 
sound of A in Copenhagen. It is 
sometimes heard in 19., especially 
from ladies, as a thinner utterance 
of (aa) than (aah) would be. 

(^u seu). See (eu). 

(Ah ah). This thin sound is seldom 
heard in 19., except in the pronun- 
ciation of delicate speakers, in such 
words as : basket, staflF, path, pass, 
aunt, and, as Mr. M. Bell believes, 
for the unaccented a in amount, 
canary, idea, and rapidly pronounced 
and. It is also the first element in 
the diphthongs : high, hoiv, as pro- 
nounced by some (nahi, nahu) in 
place of (H9i, h8u). It may have 
been the transition sound between (a) 
of 16., and (ae) of 17. It has the 
same representatives as (aa, a), gene- 
rally a, sometimes au. 

(Ahi ahi). See (ai, ai). 

(Ahu ahu). See (au, 8u). 

(Ai ai), if this diphthong occurred at 
all in ags. it was represented by aj, 
and seems to be the a^^^ of Orrmin. 
In 13. it was written ei, ey, ai, ay, and 
this representation continued, per- 
haps, through 16. After 16. the 
sound seems to have disappeared, 
but probably remained in a few 
words, and in 19, it is generally 
heard in the affirmative ay, or aye, 
and from many clergymen in lsa«ah. 
In the provinces it is a common 

pronunciation of long t. Mr. M. 
Bell considers that sound, however, 
and the German pronunciation of 
ei, ai, to be (ai), and (ai) to be the 
general sound of English long i ; in 
that case (ai) would then have the 
expressions given below for (ai). 

(Aa aA), this French sound has only 
recently been introduced into Eng- 
lish, but is firmly established in aide- 
de-camp {ee di kaA), the last word 
being called (kAAq, koq, kaemp) by 
diff"erent orthoepists, but (kAq, koq) 
would not be endured, and (kon) is 
more often said. In ^wvirons (aA*- 
viroA, envaiiTonz), an envelope 
(aA^vikp, en-velap), custom varies. 
For e7inm. the pronunciation (aAwii-), 
or (onwii), is common, (oqwii*) is 
passe, the old form was annoy, = 
(anui*). Perhaps it would be more 
correctly written (aa) as pronounced 
by Englishmen, the labialisation 
being disclaimed by Frenchmen. 

(Au au), in Orrmin awtv, in 14. to 16. 
au, aw. This sound was lost in 17. 
and has not been recovered, though 
somedeclaimers still say (aul) for (aaI) 
all. Heard in the provinces. It is 
the German sound of au. Mr. M. 
Bell, however, considering this last 
to be (au), and believes (au) to be the 
usual sound here assumed to be (au), 
in which case it would really exist 
in the language, and be expressed as 
(au) is stated to be below. 

(B b), always expressed by b, or bb. 
The mute final e, and assimilated 
letters, have produced the 19. va- 
rieties : ^e, Qbb, ebbed, habe, Cock- 
bnrn (Koo'bin), HoMorn, cujoioard 
(this was also in 17.), haxitboy (hoo*- 
boi). In 17. Jones finds deputy, 
cnpid, etc., pronounced with (b). 

(Bh bh). It is doubtful whether this 
sound was ever known in England, 
but Dr. Bapp considers it was ags. 
tv. It is possible that the southern 
(London and Kent) tendency to con- 
vert (v) into (w) may arise from 
some original mispronunciation of v 
as (bh). The sound is not only not 
acknowledged, but is rarely under- 
stood by Englishmen. Even in parts 
of North Germany (bh) has been re- 
placed by (v). See the description 
of the sound, p. 513, note 2. 

(D d), always expressed by d, dd. The 
mute final e, and assimilated letters, 
together with foreign words, have 
produced the 19. varieties: bdellmm, 

Chap. VI. § 2. EXPRESSION OF SOUNDS. (Dh— E). 


<feep, didd^ Bu^f^^ist, \xa,de, "Wync?- 
Aam, \oyed, ^ould, bur^Aen, usually 
burc?en. In 17. they had: souWier, 
wou^c?, etc., bur^Aen, murder, etc. 

(Dh dh), this sound must have existed 
in ags., but it is not possible to say 
whether ]?, or ^, was meant for it. 
In Icelandic ]> is (th), and "5 (dh), 
but they must have been confused in 
ags. at an early period. See supra 
p. 515, p. 541, n. 2, p. 555^ n. 1, 
col. 2. Even Orrmin does not dis- 
tinguish them. "When th was intro- 
duced it was used indiscriminately 
for (th, dh). The 19. sign is still 
th, though there seems to be a feeling 
that e final will ensure the sound 
(dh), as breaziA, brea^Ae (breth, 
briidh). Some literary men write 
dth to indicate the sound. 

(Dj dj), an unacknowledged English 
sound, common in speech in 19., and 
represented by d before w, as : ver- 
duxe = (yrdjux), when the speaker 
wishes to avoid (vj-dzhi). It is pa- 
latalised (d), a transition sound be- 
tween (d) and (dzh), and is distinct 
from (dj). Vulgar speakers do not 
change would you ? into (we^dzh'j), 
but into (wwdj'B). Some even say 

(Dw dw). See (dt^). 

{Qw diw) is perhaps the true sound 
heard in : dweW., divaxt generally ac- 
cepted as (dw), with doubts as to 
whether it is not (du). It seems to 
be an unacknowledged lip modifica- 
tion of (d), so that (d) and (w) are 
heard simultaneously, rather than 
consecutively, the lips being rounded 
as for (w), while the tongue is raised 
for (d), and the separation of the lips 
and of the tongue from the palate 
taking place at the same time to 
admit the passage of the vowel. 
How long this sound has existed as 
distinct from (dw, du) cannot be 

(Dzh dzh), does not seem to have oc- 
curred before 13., and arose first 
from palatisation of final (g) in ags., 
which, after short accented vowels 
in closed syllables, passed through 
the form (//), rather than (^h), and 
hence generated (dzh) in place of (j), 
as : edge, hedgfe, ledge, fledge, com- 
pare ags. ecg, hege haeg, lecgan, 
nycge ; and, secondly, from the 
French i consonant, and </ before e, t, 
which there is good reason to sup- 
pose was pronounced at one time as 

(dzh), and which is said to be (dz) 
in present Provencal, by a vn-iter 
who confuses the Spanish cA, which 
is (tsh), with (ts), (Mireio, Mireille, 
poeme proven^al de Frederic Mistral, 
avec la traduction litterale en re- 
gard, 8vo., 1868, p. vii). Hence it 
is expressed by i consonant, ff, gg, 
dg. Subsequently only /, g, dg 
(the latter before e generally) were 
used, but not consistently. In 19. 
we have : GreenwicA, soldier, which 
was also heard in 17., with omitted 
/, as (soo'dzher, SAdzh-er), y^dg- 
ment, ridge, W^<5?wesbury (Wedzh'- 
beri), ^em, colley^, Belling Aam 
(Belindzhem), /ust. 
(E e), this, or (e) was the ags. short 
e, and has prevailed in one form or 
the other to this day. I am myself 
in the habit of saying (e), but this ap- 
pears too delicate to Mr. Melville 
Bell, who prefers (e), which is 
the Scotch sound, and is in Scot- 
land by many English people con- 
fused with (ae), see p. 271. It was 
occasionally expressed by u from 13. 
to 16. Being an exceedingly com- 
mon sound, it easily absorbed related 
sounds, and hence even in 17. had 
numerous forms of expression, the 
only normal form both then and 
now being e, but ea was very common 
in 17. as in 18. and 19., and ai in 
17. in unaccented syllables as cap- 
tam, now (kaep-t/yn), nearly (ksep'ten) 
or (kaep'tm). Before r it seems to 
have been the refuge of other sounds, 
which however may be more pro- 
perly ('b). The following are 19, 
varieties : m«ny, Pontefr«ct (Pom-- 
fret), Pcestum, Mich«d, Thames, 
said, Abergawnny [(JEb-jgene) vnrit- 
ten Ahurgany in the Shakspere folio 
1623, Hen. viii. i, 1, speech 49, where 
it must be in four syllables for the 
metre ; this is not the "Welsh pro- 
nunciation, but is common in Eng- 
land,] s«ys, let, head, de5t, Wednes- 
day, allege, foreAead, het'fer, Leicester, 
leopard, cheqwe, rendezvous, rAetoric, 
freend, consce'ence, foetid, connojsseiir, 
bwry, gwess, paneg«/ric, [this pro- 
nunciation is going out, as also that 
in spirit, s?/rrup, ste'rrup], gunz^ale, 
Thomas's (I'om-'Bsez). If the sound 
is admitted in the syllable (ei) for 
(i) then we might add : sabre, virtue, 
Bridlington, sapph?re, bettor, Ur- 
qwAwrt, anse^er. Most of these ex- 
pressions are highly exceptional, and 


EXPRESSION OF SOUNDS. (^&— a^). Chap. VI. ^ 2. 

arise partly from assimilations and 
omissions, and partly from inser- 
tions. Still the spelling has re- 
mained and has to be separately 
memorized by those who would use 
it, as no rule can be assigned. 

{E ey It is impossible to say whether 
this sound occurred in ags. or old 
English as distinct from (e). Whe- 
ther the final unaccented e of 13. and 
14. had the sound of {e) or (e), or 
whether it was not rather (i?), is also 
impossible to determine. In 19. the 
sound only occurs as short and un- 
accented, in some words, as aerial^ 
aorta (eer'iBl, ^oi'ta), for which some 
would read (ahoi-tah). It is the 
French e. 

(:E e). This is a variety of (e) and in 
the pronunciation of some persons 
imiformly replaces it, and has been 
therefore always expressed as (e) was, 
wherever it occurred. See (e) . 

(a 9)* This sound does not appear in 
English till the middle of 17. It 
is not named by Butler, 1631. It 
is distinctly recognised by Wallis, 
1653, and Wilkins, 1668, and all 
subsequent writers. It replaced (u) 
and was expressed as (u) had been by 
M, o and occasionally om, and these 
have remained its principal forms 
to 19., but numerous degradations 
have occurred especially in unac- 
cented syllables, where, however, 
stricter analysis seems to shew that 
the sound is now rather (-b) . Thus 
we have the 19. varieties : riband, 
meerschawm, escutcheon, /mmble, 
vaotion, consceoMS, son, does, love, tor- 
toese, Lincoln, flood, do?<ble, tongwe, 
bello^t's, tt<;oppence (in 17), — and if 
we consider that (.i) is properly (ai) 
we have this vowel in : amatewr, 
cupboard, avoudupoise, colonel, 
liqueur, liqwor. Mr. M. Bell uses 
(a) for (e). 

(sr ^)- This French sound should of 
course be used in those French words 
containing it, which are used in 
English, but it is always replaced 
by the familiar (g, .i). 

(^ v). This faintly-characterised vowel 
is recognized by Mr. Melville Bell 
as the real sound in unaccented syl- 
lables, where 19. orthoepists usually 
assume (a, a) to exist, before >?, /, r, 
and 5, as : motion ocean, principal, 
Tartar, facetwws. It is therefore 
expressed by any combination de- 
noting unaccented (o, ae). 

(Ee ee.) In earlier English down to 

18. we cannot distinguish (ee, ee). 
In ags. it seems to have been re- 
presented only by e or e. In 13. it 
was also represented by cb, and oc- 
casionally by ea, eo, at least, these 
forms all interchange with e. In 14. 
eo was almost quite dropped (though 
both eo, oe are occasionally found), 
and ea was very sparingly used, but 
ee was common, especially in closed 
syllables. In 16. the practice was 
introduced of representing (ee) by 
e, ea only, to the exclusion of ee. 
During 17. at, ay, ei, ey were used 
as well as e, ea, but the two latter 
forms were less and less used as (ee), 
till they became exceptional expres- 
sions in 18. and 19. In the middle 
of 18. the usual forms were a (with 
any addition which, shewed prolon- 
gation, as a final mute e), ai, ay, 
occasionally ea, and ei, ey, but the 
two last forms were rapidly going 
out, and at the end of 18. and be- 
ginning of 19. few remained. In 
19., if not earlier, (ee) was separated 
from (ee), and the sound of (ee) was 
only used before r (j) , but it was ex- 
pressed by all the same forms as (ee). 
This limitation of the sound of (ee) 
reduces the number of its forms in 

19. where we find: ^aron, mare, 
aerie, air, Ayr, mayor, -pear, ere, e'er, 
their, eyre, heir. See (ee). 

{He ee). This sound was not consciously 
separated from (ee) till the end of 
18. or till 19. Even now many 
persons do not perceive.the difi'erence 
(ee, ee), or if they do hear the sounds 
they analyse them as (eei, ee). In 
some parts of England (ee) alone is 
said, in the South many people can- 
not pronounce (ee) before any letter 
but (a), and cannot prolong (ee) 
without dropping into (i), thus (eei). 
Some assert that (ee) is never pro- 
nounced, but only (eei), with which 
they would write the words : mate, 
champagne, dahlia., -pain, campaiyn, 
straight, trait, halfpennj, often 
(naa-peni) in the North, gaol, Cars- 
halton (kees'HAAt'n), gauge, plague, 
play, great, eh ! \eil, reiyn, weigh, 
they, eyot. 

(g[a 89). Never a recognized sound, 
but one from which (jj) is with 
difficulty distinguished. It is there- 
fore heard in place of (oj, ea), or 
rather (ai, u.i), by the representatives 
of which it is always expressed. 

Chap. VI. § 2. EXPRESSION OF SOUNDS. (Eei— F). 


(Eei eei.) In 16. Gill acknowledges 
(eei) and frequently writes it in the 
word they (dheei). It probably ex- 
isted in 17., as it is partially ac- 
knowledged by Cooper. If so it was 
written ee, ey^ ai, ay. Most probably 
its use increased in 18., but there is 
no proper note of it. 

{Eei eei.) This sound is not acknow- 
ledged before 19., and then the ex- 
tent of it is disputed. Some make it 
coextensive with the spelling ai, ay, 
others make it replace the sound of 
{e.e) under whatever form it is ex- 
pressed. Some persons in the South 
of England seem incapable of sus- 
taining (ee) or (ee) without rapidly 
falling into (e, i). See {ee). 

(a[h oh.) This replaces (o) under what- 
ever form it may be expressed, in the 
pronunciation of many persons. It 
is the form acknowledged by Mr. M. 

(Ei ei.) In 16. this is acknowledged 
by Salesbury, and Hart as the sound 
of i long and of ei, ey. Smith ac- 
knowledges it in a few words, con- 
taining ei, ey, where he doubtfully 
distinguishes it from (ai), but he 
marks i long as a separate vowel, 
which he identifies with the English 
words for ''ego, oculus, etiam," I, 
eye, aye. Gill sometimes writes (ei), 
sometimes (eei), in the same words, 
and considers long i to be very nearly 
the same. Wallis does not acknow- 
ledge the sound, and it seems to have 
expired in 17. It is, however, re- 
viving, although unacknowledged, as 
a substitute for (eei) and that for 
{ee), as (rein) rain. 

{Ei ei.) A variant of (ei), which 
cannot be properly distinguished 
from it in accounts of pronunciation, 
but seems to be the true sound of 
the modern Scotch long i in many 
words, see p. 290. 

(gi oi), or perhaps (oi) is acknowledged 
by Wallis and Wilkins in 17., and 
was perhaps intended by Gill as the 
sound of long i, and has since re- 
mained that sound, though individu- 
ally and provincially replaced by (ai, 
ahi, ei, ei), etc., see p. 108. It is 
expressed by any combination of 
sounds which indicate that i or y is 
to be long. Hence in 19. we have : 
naiye, aisie, de«pnosophist (and as 
many pronounce e2ther, netther) 
"height, the older sounds (neet, neeit) 
are occasionally heard, (nekht) is 

still heard in Scotland, (nekth) has 
been noted in the neighbourhood of 
Ledbury, Herefordshire, (naitth, 
hoith) are mistaken pronunciations — 
eyiu^, eye, rAmoceros, E^me, xhyvo.- 
ing, rhyme, bmd — this mode of ex- 
pressing long i is found as early as 
16., — indict, die, live, sign, siy/i, 
sighed, vtscount, isle, heguiling, he- 
guile, huy, %, dye, seyihe. 

(Ea ca) is not an English sound, and 
no attempt to pronounce it occurs 
before 18. In 19. coup de main, 
which Feline writes (ku-d mcA), is 
written (kuu-dimaeq) by Worcester, 
(kuu dima3q-) by Webster, (kuu'de- 
maaq) by Knowles, (kuu-damseq:) by 
Smart, (kuu-dimem) by Mavor. It 
is generally called (kuu'd? mcA), 
though some affect the complete 
French pronunciation. 

(a[A oa), this is also not an English 
sound and is so rare in French that 
it is seldom borrowed in English, 
except in the name of the game 
vingt et un, usually called (vcAtoA-) 
in England, often corrupted to (vaen- 
tiun*, vaendzhon*), just as rouge et 
noir becomes Russian tvar, from the 
older pronunciation, still occasionally 
heard, of (Ruu'shen wAAr). 

(Eu eu) Common in 13. and 14. as the 
sound of eu ew, from ags. eaw, etc. 
Less frequent in 16., expiring in 17., 
and lost in 18. In 19. it is frequent 
as a London pronunciation of (au), 
thus (deun teun) for down town, and 
either in this form or (eu, a3u) com- 
mon in Yankee speech, and in the 
East Anglican dialect. It is acknow- 
ledged in Italian and Spanish ^wropa, 
and in modern Provencal, both eu, 
and ie'u (eu, ie'u) are distinguished, 
the last word being the French/^. 

(;Eu, eu). See (eu). 

(a^u ou). Not known before 17. In 
17. and since, acknowledged as the 
sound heard in now how, though some 
pronounce (au, ou, ou, au, ahu) and 
even (a3u, eu). Expressed generally 
by ou, ow, with or without mute 
letters. In 19 we find: moMtchouc, 
Mackod, hour, compter, noun, doubt, 
renounce, hough, cow, allow^^d. 

(F f). From ags. to present day re- 
presented by /, ph, with their dupli- 
cations ff, pph. From 16., at least, 
occasionally expressed by gh. In 19. 
we find : foe, fi/*?, sti/; stu/i;d,/ugle- 
man — a mere corruption — o/iJen, 
laugh, half, sapphire, lieutenant. 



(G g-). From ags. to present day ex- 
pressed by ff. In 14. also by ff^ and 
in 15. also by gge final. Ghost is 
found in 16. In 19. we have blac^- 
ffu&Tdy go, egg^ be^^ed, ^Aost, guess, 

{G g) or (gj), palatalized (g). Probably 
in ags. g before a palatal vowel, subse- 
quently (dzh). After that change {g) 
cannot be clearly traced before 18., 
but it is still found in 19., represented 
by g, gu, before a (aa, ai) or long i 
(ai), as : garden, y ward, regard, ^wide. 
In 18., it seems to have been also 
used before short a (se). 

(Gh gh). In ags. perhaps more cer- 
tainly in 13., expressed by 3, after 
a, 0, u long and followed by a vowel 
as o-^en. Possibly the sound after 
0, u was labialized to (gi^^h). Whether 
these sounds were entirely lost in 14., 
being replaced by (kh, kwh), it is 
difficult to say ; probably not. As 
long as they lasted they were ex- 
pressed by 2, gh. It must have been 
lost in 16. 

{Gh. gh). In ags. perhaps, more cer- 
tainly in 13., expressed by 3 after 
Cj i long or short, and occasionally 
after r, l, in which case it fell into 
{%). In ags. perhaps the initial 
sound of 3 before palatals, which in 
13. was replaced by (j). In 13. 
written 3, 3 A, yh. After 13. gene- 
rally replaced by (^h, j), and written 

(Grh grh). Only known as a local 
peculiarity, the Northumbrian burr, 
and then expressed by r, rr as in 
Harriet (Hagrh'iot). See (r). 

{Qw gw). The labial modification of 
^, confused with (gw), from which 
it diflfers almost as simultaneity from 
succession, {^w) resulting from at- 
tempting to pronounce (g) and (w) 
at the same time. How long it has 
been known in English cannot be 
determined, but it is probably a very 
early combination in the Romance 
languages. In 19. it is expressed by 
gu in : ^e<aiacum, ^wano, ^wava 
(gt^ai-akam, gwaa-no, gw^aa'va). 

(Gwh givh). Probably an ags. sound 
of 3 after labials, and occasionally 
r, I, in which case it became (u, 0). 
In 14. probably expressed by gh 
after 0, u. Perhaps lahh, laugh, 
lauwh, indicated (lagh, laugwh, 
lauwh) passing to (lauf). But the 
sounds may have been (lakh, laukw^h, 

(H h). The true aspirate consisting 
of a jerked emission of the following 
vowel without the previous inter- 
vention of the whisper, was, proba- 
bly, the genuine old form of aspira- 
tion, as shewn in the Sanscrit post- 
aspirates. It was frequently inter- 
changed with (h', kh, gh), the last 
(gh) being the value of the Sanscrit 
^ usually considered as h. Repre- 
sented whenever it occurred from 
ags. to 19., by h. See (h'). 

(H* h'). The jerked utterance accom- 
panied by a whispered breath pro- 
ceeding the vowel. The jerk is of 
importance; (*a-aa), is difl'erent from 
(H-*a-aa=H'aa). Constantly occur- 
ring, and represented by A, but in 
16. occasionally by gh. In 19., 
either (h) or (h*) according to a 
speaker's habits of utterance, and fre- 
quently according to the momentary 
impulse of the speaker, is expressed 
by the following varieties : Calla^^Aan 
— and by gh in many other Irish 
names — hole, Colquhoun, whole. 
Uneducated speakers, especially when 
nervous, and anxious not to leave 
out an A, or when emphatic, intro- 
duce a marked (h') in places where 
it is not acknowledged in writing or 
in educated speech. On the other 
hand both (h, h',) are frequently 
omitted, by a much more educated 
class than those who insert (h'), and 
in the provinces and among persons 
below the middle-class in London, 
the use and non-use of (h, h') varies 
ft'om indi\ddual to individual, and 
has no apparent connection with the 
writing. Hence its pronunciation 
has become in recent times a sort of 
social shibboleth. The very uncer- 
tain and confused use of h in old 
MSS., especially of 13., serve to 
make it probable that there was 
always much uncertainty in the pro- 
nunciation of h in our provinces. 
The Scotch never omit or insert 
it, except in hicz (naz), the emphatic 
form oius. The Germans are equally 
strict. But the sound (h) or (h') is 
unknown in French, Italian, Spanish, 
modern Greek, and the Sclavonic 

(I i). Whether this sound existed in 
closed accented syllables before 16., 
is doubtful, probably not. After 16. 
there is reason to suppose that if it 
did exist, its use must have been 

Chap. VI. § 2. EXPRESSION OF SOUNDS. (^— Tu). 


very limited. In Scotland it both 
did and does exist. In all cases it 
was represented by «, y. ks, a short 
sound in open syllables it was pro- 
bably quite common, and was in ags. 
to 14. represented by i. In 16. this 
short open (i) was ^ as in : beleeve 
• (biliiv). At present the distinction 
between (i, i) in such cases is rather 
doubtful, and both are apt to be 
merged into (■b). But where the 
distinction is made, short (i) is 
always expressed by e ; see {%). 

(J i). This seems to have been the 
common sound represented by short 
i in close accented syllables in 
ags., and by short ^■, y^ and' occasion- 
ally u in this situation from 13. 
to 19., and with tolerable certainty 
from 14. to 19. In 16., as a final, 
it was frequently written ie. Or- 
thoepists, however, constantly con- 
fuse (i, i) both in closed and open 
syllables, so that any real separation 
of (i, i), is hazardous. In 19., (e) 
in closed syllables is expressed in a 
great variety of ways, owing to 
various degradations, but generally 
as i, y with some letters which have 
become mute, and when in final 
open syllables, generally by y or 
some variety of the same. The fol- 
lowing forms may be noticed. In 
closed syllables : landscape, Samt 
John (Sm'dzhBn) as a family name, 
Jervawlx (Jaaj'Vis), pr(?tty, guinms, 
beau^n, bre(?ches, forfett, Th(?obald 
(Tih'vld) the recognized name of an 
editor of Shakspere and a street 
in London, honsewife (H^z•^f) a 
threadholder, exhih'it (egzib'it) 
some say (ecsH'^■b•^t) with a very 
marked (h'), vhythm, ptt, mar- 
liages, marriage, pit^Vd, to liye, 
sieye, iivej^ence, women, groats 
(gr«ts), Jervois, Misi;r<?ss (M^;s•^■s), 
bMsy, lett?<ce, huild, bwsmess, Tyr- 
whitt {Tirit), Ciiiswick {Tshiz-ik), 
physic, Wymondh.a.m. (Wind'Tsm). 
In open syllables, many of the above 
forms and : Rothsay, money, Annie, 
Beaxdieu (Biu-h"), felloi? (fel't), 
chsanois leather (sh8em•^■), plagw?/. 

(li ii). In ags. either (ii) or (n), 
which see, was always expressed by 
i long, and so on to 14. and part of 
15. After 15. (ii) was only rarely 
expressed by i long, but more and 
more frequently by e, ee, and in 16. 
frequently by e ee and rarely by ea, 
ie. The expression by ea, ie increased 

slightly in 17. In 18. e, ee, ea, ie, 
were the rule, and ei, ey the excep- 
tions. In 19. the two latter also 
became the rule. The Latin ae, oe 
werer also added to the list, and vari- 
ous degradations swelled the expres- 
sions of (ii) in 19. to the following 
extraordinary variety : minutice, de- 
main, Caius College, he, each, ^eaed, 
\eaye, Bm^<champ (Bee'tsh^m), 
\eague, ieei, e'en, compete, sleeye, 
impre^n, jjegh, conceit, conceiye, 
seigniory, heigh, receipt, Be^voir, 
p(?ople, demesne, ]i.ey, Wemyss 
(Wiimz), keyed, diarrAcea, invalid, 
grief, magazme, grieye, s?^niour, 
fusil, debrt\$, intrigue, fojtus, quay, 
quayed, mosquito, turqtioise (tjkiiz*) 
according to Walker, Smart, and 
"Worcester, more commonly (tyr- 


{Ii ii). In 14., and most probably 
earlier, the sound of long i and y. 
During 15. this sound nearly ex- 
pired and was only retained by a 
few individuals in 16., being re- 
placed by (ei, i) according as the 
syllable in which it occurred retained 
or lost the accent. It is heard in 
Scotch in 19., where a short {i) is 
accidentally lengthened as : gi'e, wi'. 
In English it is an unacknowledged 
sound often heard from singers who 
lengthen a short {i), as {stiil) for 
(stiV) still, as distinct from (stiil) 
steal, see pp. 106, 271. 

(lu iu iuu). These sounds cannot well 
be separated. 'I'hey probably never 
occurred initially. When Smith 
wrote iunker in 16. he meant (juq*- 
ker). The sound was not recog- 
nized till 17., when it was generally 
expressed by long u, or eu, ew. The 
same combinations used initially, as 
in use, unite, ewe, probably expressed 
(jiuu, jiu, jiuu). In my phonetic 
spelling I have seldom thought it 
necessary to distinguish (iu, iuu) 
and have frequently omitted to pre- 
fix the (j). From these sounds 
should be distinguished (juu, ju) 
which are also confounded with 
them, but are usually written you. 
With these the sounds (jhiu, jhiuu) 
often confounded with them, had 
best be considered. The following 
are the 19. varieties of expressing 
these sounds : 
(iu) monument, document, incwbate, 

(iuu) heautj, feod, feudal, deuce. 


EXPRESSION OF SOUNDS. (J— Lh). Chap. VI. § 2. 

"Levesorif new, adieu, view, viewed, 
f/'Wglcman, ame^sing, fuchsia, (fiuu*- 
shia), cue, amuse, queue, impi<^n, 
huh\, suit, -puisne, (piuu*nj), lute- 
string (liuu'stri'q,) fugue. 
(jiu) tinitc, Eugene (Jiudzhiin*) 
(jiuu) euffh, ewe, yew, lyule. 
(ju) in 16. young = (juq) like present 

German jur/f/. 
(juu) 1/ou, 1/outh. 
(jhiu) ^e^mane. 

(jhiuu) htcman, hue, Hugh, Hughes. 
(J j). The palatal consonant into 
which ags. initial {g\\) degenerated, 
generally confounded with an initial 
unaccented (i), whence it is occa- 
sionally derived, and often confused 
with the palatal modification (j) 
from which it differs as (w) from 
{w). Apparently in use from 13. to 
19., expressed in 13. and often in 
14. by 3, ), whence the modern forms 
y, z, p. 310, and p. 298, note. The 
varieties in 19., are : hideous, oni'on, 
hallelujah, yard, Denzil. 
(Jh jh). Orrmin's jh in iheo she. 
The whispered (jh) differs from (j), 
as (/fh) from (^h), but is by Germans 
confounded with (A-h), although often 
pronounced by them quite distinctly 
in ja {3\xaa) for [iaa). It has pro- 
bably often been pronounced in 
English, but it is not recognized, 
and even in the words cited under 
(iu) it is not now generally acknow- 
ledged, (jhiuu) being taken as (h'juu, 
H'iuu) sounds which are not easy to 
Titter. It has no special representa- 
tive, but is implied by any combina- 
tion apparently expressing (n'+iu). 
(K k). The sound has been in use 
from ags. to 19. In ags. expressed 
by c invariably. In 13. generally 
by c, occasionally by k. In 14. by k 
and occasionally by kk, ck, but fre- 
quently in words from the Latin and 
French by c, cc. In 16. by c, cc, k. 
ck, and occasionally ch. In 17- gh, 
qu were added to the list. All these 
remain, except kk, which was dis- 
used before 16. In 19. we have : 
can, account, Bacc/mnal, scAool, ache, 
'hack, backed, ac^^uaint, hongh, kale, 
hake, 'walk, ^uack, quay, antique, 
TJrquhart, viscount, hatchel (H8ek''l) 
also written hackle, heckle, e^rcept. 
(K k). This is the palatalized form 
of (k), see g, and its existence was 
acknowledged, and expressed in 18. 
by c, k before a (aa, ai, ae) and i (ai) 
as in: cart, candle, s^y. This is 

regarded as antiquated in 19. but is 
still heard. 

(Kh kh). In ags. expressed by h, hh\ 
in 13. by ^, gh, and very rarely by ch, 
p. 441, from 14. to 16. by gh. After 
16. lost in English, though common 
in Scotch, where it is usually written 
ch. At no time were the palatal 
and labial modifications (kjh, kt^^h) 
distinguished in writing from (kh), 
but there seems reason to suppose 
that a preceding vowel when palatal 
determined (/ch = kjh), when gut- 
tural (kh) and when labial (kt<;h). 
See also (gtt'h). 

(Xh ^h). See (kh). 

\K.iv Vw). This sound has always 
been confused with (kw), but there 
is reason to suppose that (kw) has 
been the real sound from the earliest 
times, pp. 512, 514, 561. In ags. 
(kw) was expressed by cw, in 13. qu 
seems to have been introduced and 
to have remained to 19. 

(Kwh kwh). See (kh). 

(L 1). From ags. to 19. I and from 
14. to 19. U is frequent. In 19. 
mute letters have occasioned the fol- 
lowing varieties : sera^/io, maA/stick, 
/ace, Gui/f/ford, ale, ill, trave/M, 
kiln, isle, hxistly, viGtiiallev (vet-b). 

('L '1). In 16. certainly, this sound 
was expressed by final -le forming a 
syllable, and it was recognized by 
Bullokar after a and before another 
consonant, as haM (na'lm) where 
others read (ul). In 19. several 
phonetic writers incline to {u\), but 
the majority consider (1) only, to 
be the sound. Mr. M. Bell considers 
it to be (11) that is lengthened (1). 
It is always represented by -le or -l. 
It generally falls into {I) when a 
vowel follows as double doubling 
(dab'l dab-h'q), but some persons re- 
tain the (') and say double-ing (dab*- 

(Lh lh) . Not now a recognized Eng- 
lish sound, but it occasionally arises 
when instead of prolonging an (/) 
with the full murmur, the action of 
the vocal ligaments ceases, while the 
tongue remains in position, and the 
unvocalized breath escapes on both 
sides as (fAAllh). It is also recog- 
nized by Mr. M. Bell in felt (fElht) 
or perhaps (fEllht), as he would 
write. In Modern French it is very 
common for (1') as (tablh) table, and 
hence it has been recently imported 
into the English pronunciation of 

Chap. VI. § 2. EXPRESSION OF SOUNDS. (Lhh— (E). 


French words. It was probably the 
sound written hi in ags. and Ih in 
13., as it is now represented by M in 
(Lhh Ihh). Few Englishmen can pro- 
nounce this Welsh sound properly, 
but as Welsh names of places are 
current in English, as Llangollen 
(Lhhangolhh-en) it should be recog- 
nized, and not treated as (thl) or 
(tl), as in (Thlangoth-len). For a 
description of the sound see Chap. 
VIII, § 1, under II. 

(Lj, Ij). An unrecognized English 
element, often generated in the pas- 
sage from (1) to (j) or (i) before 
another vowel. Thus million, bul- 
lion are rather (m^l•ljBn, bwMJTjn) 
than pure {mil-jvn, hiA-j'Bn) because 
there is no break, thus (1,j), but the 
(1) is continued on to the (j) pro- 
ducing (lj =l*j). Some Englishmen 
pronounce seraglio, lieu, lute, as 
(seraa'ljio, Ijiuu, Ijiuut) others say 
(seraa'lio, luu, luut). 

(M m). From ags. to 19. m, and 
from 14. often mm. In 19. we have 
the varieties, chiefly assimilations 
and degradations : drac/?;w, phler/m, 
psa^m, Cho/mo>^6?eley (Tshamij), aw, 
\di.mb, tame, hammer, shammed, 
hymn, Campbell (Ksem-el), BanS 
(Baemf), Powz^efract (Pom-fret). 

('M 'm). Certainly from 16. when it 
was recognized by Bullokar. Not 
distinguished from (m) in writing, 
and not recognized as a syllable in 
poetry, as : schism, rhythm (s«z"'m, 

(Mh mh). Eecognized by Mr. Mel- 
ville Bell in 19. before j?, t, as lamp, 
ewpt (Isemhp, Emht) or (laemmhp, 
Emmht) . 

(N n). From ags. to 19. n and from 
14. nn. Silent letters and assimila- 
tions, etc., have produced the 19. 
varieties : stnddinffsail (stan-s'l), 
op^wing, ffnaw, John, k?iOW, Coln- 
brook (Koon-hrtck), Galne (Kaan), 
mnemonies, compter, can, rihand, 
cane, ipecacua?2Aa, manner, plawwed, 
gnnwale (ggn-el), reaso^dng, pneu- 
matics, puis;^e (piuu-ni). 

('N 'n). Certainly since 16., repre- 
sented by -e7i, -on, as in : opew, 
reasow. When a vowel follows the 
(') is lost, though some say (l8it"'n«q) 
and others (bit'neq) lightening, light- 

(Nh nh). Eecognized in 19. by Mr. 

M. Bell in tent, which he writes 
(tEuht) or (tEnnht). 

(Nj nj). An unrecognized English 
sound produced by continuing the 
sound of (n) on to a following (j, i) 
as onion, more properly (an-nj^n) 
than (an*,JBn). Some call new 
(njiuu), others (nuu). Common 
French and Italian gn. 

(0 o). This seems to have been the 
original ags. and English short o up 
to 16., and to have been lost, except 
in the provinces, after the middle of 
17. when it was replaced by (a, o). 
It is the French hommage (omazh) 
as distinguished from 19. homage 
(Hom-ydzh). It is Italian short o 
aperto. It is also heard in Spain, 
Wales, and a great part of Germany, 
though it is liable to fall into (o) 
on one side and (o) on the other. 
In old English invariably o. 

{0 o). This short sound in closed sylla- 
bles is not recognised in 19., but it 
is heard the provinces and in America 
for short and sometimes long o ; thus, 
whole stone [uol, ston), and then is 
scarcely distinguishable from {u) or 
(a), and is confounded by some with 
(a). In open syllables it is not un- 
common, as in : oblige, memory, win- 
doiv (oblaidzh*, memor*', wm'do), 
where it is often confused with (a, v), 
and even, when final, with (j). It, 
probably, came into use with (oo) in 
17., but was not distinguished from 
it. Generally expressed by o, ow, as 
above, and in 19. we call Pharaoh 

(q o). In 17. short o passed from (o) 
to (a) or (o). The distinction be- 
tween these sounds being of the 
same degree of delicacy as that be- 
tween (i, i) and (e, se) renders it 
difficult to determine which sound 
was said. In 19. (o) prevails, though 
(a) is occasionally heard, and may 
be heard when the expression is «;, 
au, or (a) influenced by (u) in any 
way. See (a). The general ex- 
pression of (o) is ; but in 19. we 
have the varieties : resvn, hononr, on, 
groat, forehead, cognisant, JoAn, 
ho2^gh, pedagog?<<?, knoi^^ledge. In 
or not followed by a vowel, the theo- 
retical sound is (oj), the actual sound 
scarcely distinguishable from, if not 
identical with (aa', aa). See supra 
p. 575, under o. 
((E oe) is not a recognized English 
sound, but is heard in the provinces 



Chap VI. § 2. 

and in Scotland, and written o, oo. 
Confused in English with (q). 

(^ oo). Recognized in 19. by Mr. M. 
Bell as the vowel in : prefer, earnest, 
ftVra, myrrh, ^uer^on, where he 
writes (ooa) for the italicized letters. 
I do not distinguish these sounds 
from (^), and in general find them 
confused with (a). See these sounds. 

(CEoc oece). Occurs in the provinces, 
and probably in Scotch. It is the 
German long oCy as in Goethe 

(Oi oi). With this must be taken (Ai, 
oi, oi ; Ae, ot, oi). It is very difficult 
to determine the limits of these 
sounds in time or place. Probably 
in 16. when oi, oy were not (ui), 
they were (oi). In 19. (Ai, ot) pre- 
vail, (oi, ui) are provincial. The 
expression is always oi, oy, with 
or without some additional mute 
letters. In 19. we have: bourg^o/s 
(bidzho/s-) noisy, noise, poe^nant, 
coigne, hoy, enjo?/ed, ^oy\e, (\uoii ; 
some say (kivdit), huoy ; some say 
{htvoi), buu/), huoyed. 

(Oo 00 ). From ags. to 16. this was 
the recognized long sound of o, and 
expressed by o, oo. It is still heard 
in the provinces. It was apparently 
lost in the received dialect in 17., 
but revived in 19. before (i), as in : 
oar, ore, o'er, moor, mourn, "pour, four, 
sword. Sometimes heard before /, s, 
th, as : ofl-, cross, broth (oof, croos, 
brooth), where it is apt to degenerate 
into (aa, oo), or sink into (o). 

(Oo oo). From 17. the recognized 
sound of long, and generally re- 
presented by 0, o-e, oa, and occasion- 
ally by oe, ou, ow. In 19. we have 
the varieties : h-auienr, h.auihoy 
(Hoo•b^^), heau, yeoman, &kew, now 
frequently written show^ sewed, fre- 
quently written sowed, post, oats, 
provincially (wats), Soam^, boats- 
wain (boo's^n), Coc^-burn (Koo'bxn), 
do^, bon^, o^lio, oh, scntoire (skru- 
tooj-), according to Sheridan, Walker, 
etc., now generally (skrutwoi*), yolk, 
brooch, apropos, Grosv^nor, depo^, 
somI, rogue, Yotighall (Joo-haaI), 
thotcgh, knoM;, toM^ards, owe, Knowles, 
quoth, (kooth) ; some say (kwooth). 
See (oou). 

(qo oo). The drawl of short (o) is 
only heard in drawling utterance, as 
(ood) for (od) odd, as distinct from 
awed. Preachers often say (Good), 
but seldom or ever (GAAd) for God. 

In America some say either (doog, 
looq), or (doog, looq) for dog, long, 
etc., which the phonetic writers there 
recognize as (uAAg, lAAq), and the 
two sounds are difficult to separate. 

(Oa oa). This present French nasal is 
in older English represented by 
(uun), as retained in our modern 
balloon. In recently imported French 
words the (oa) is intended to be re- 
tained, together with its French 
expression, as bonbons, bon mot, on dit 
(boAboAZ, boA mo, oa dii). But 
the usual substitutes are (on, oq), and 
occasionally (oon, on). 

(Oou oou). From 13. to 16. the pro- 
nunciation of those ou, ow, which 
represented an ags. dw, ow. Lost 
in 17. 

(Oou oou). From 17. to 19. the usual 
pronunciation of those ou, ow which 
represent an ags. dw, 6w. This pro- 
nunciation has been, however, gene- 
rally ignored, or, if recognized, 
reprobated by orthoepists. Some 
speakers distinguish no, know, as 
(noo, noou), orthoepists generally 
confUse them as (noo), compare the 
list of words under {oo) ; others 
again confuse them as (noou). Mr. 
M. Bell states that every long o is 
(ou), meaning the same as I mean 
by (oou). Some Englishmen say 
that it is not possible to lengthen (o) 
without adding (u), and pronounce 
nearly (ou, ouu). 

(Ou ou). In 16. the general sound of 
ou, replacing the previous (uu) which 
however was heard contempora- 
neously through the greater part of 
16. In 17. the sound was recognized 
as (ou), and the sound (ou) was lost. 

(Ou ou). The modern provincial sub- 
stitute for (ou), not recognized. 

(qu ou). In 18. orthoepists recog- 
nized ow as having the sound (ou) 
or (au). It was probably an erro- 
neous analysis, which even yet oc- 
casionally prevails, owing to the 
usual orthography ou, ow. Provin- 
cially however (ou, au) may occur. 

(P p) was from ags. to 19. represented 
by p, and from 14. to 19. by pp 
also. In 19. we have the varieties, 
hiccou^A (nik'kop), pdcy, ape, Claph- 
am, tapper, flapped. 

(Q, q) was from ags. to 19. written n 
or ng, sometimes nz for nj. In 19. 
we have the varieties : ^nger, hand- 
kerchief, singer, winged, Birmiw^Aam, 
tongue, Mewjzies (M.eq'iz), p. 310. 

Chap. VI. § 2. EXPRESSION OF SOUNDS. (Qh— S). 


(Qh qh), is recognized by Mr. M. Bell 
in 19. as the sound of n before k, in 
think (thjqhk) or theqqhk) 

(E r) was from ags. to 19. represented 
by r before a vowel ; and probably 
from ags to 16. represented also by 
r even when not before a vowel. 
Perhaps lost in the latter position in 
17. Preserved pure in Scotland. 
In 19. we have the varieties : right, 
rhetoric, w^rite, hurry, catarrhal. 

('R 'r) How soon this sound came 
into English, cannot be precisely 
determined. There is reason to think 
it may have been used in 16. and 17., 
and that it generated (i). At pre- 
sent in : fearing, pairing, debarring, 
ignoring, poorer, ftery, bot^^ery, there 
is a doubt whether the sound heard 
is best expressed by ('r) or (xr). Mr. 
M. Bell gives the first, I have gene- 
rally preferred the second, see p. 

{R r). This peculiar guttural r so 
common in France and even in Ger- 
many, but unknown in Italy, seems 
to be only a softer form of the Nor- 
thumbrian burr. It is not recog- 
nized in writing as distinct from r. 

(■g; i). Probably recognized in 17. as 
well as in 18. and 19. as the peculiar 
English untrilled r, not heard before 
a vowel, and represented by final r 
together with mute letters in 19., as : 
spar^, corps, burr, mor^fgage. It 
has always a tendency to change 
preceding {ee, oo, uu) into (ee, oo, 
uu), while short a, o become (aa, oo), 
or theoretically (a, o) ; and short (t, 
e) according to Mr. M. Bell fall 
into (ao), which see. Short (a) is 
supposed to remain, as cur (koj), for 
which I prefer (ki, k'j, kia) and 
generally write (ki) as quite suffi- 
cient. In place of (i) provincially 
(aha, 91, ahi) are heard. The phy- 
siological distinction between (a) and 
(i) is very difficult to formiilate. 
There is uo doubt that in many cases 
where writers put er, ur, to imitate 
provincial utterances, there neither 
exists nor ever existed any sound 
of (r) or of (i), but the sounds 
are purely (a, a). Thus bellows in 
Norfolk is not (bel'erz) but rather 
(bEl'az). There also exists a great 
tendency among all uneducated 
speakers to introduce an (r) after 
any (a, a, a, a) sound when a vowel 
follows, as (drAA-r«q, sAAT«q) draw- 
ing^ sawing, in Norfolk, and this 

probably assisted in the delusion 
that they said (drAAa mi, saaj wwd) 
and not (drAA mii, saa wwd). In 
London : father farther, laud lord, 
stalk stork, draws drawers, are re- 
duced to (faadh-B, lAAd, stAAk, 
drAAz), even in the mouths of edu- 
cated speakers. I have usually 
written (a) final in deference to 
opinion, but I feel sure that if I had 
been noting down an unwritten dia- 
lectic form, I should frequently write 
(b, 8, a). Careful speakers say 
(faa'dh*^, lAA'd, stAA'k, drAA'z) for 
farther, lord, stork, drawers, when 
they are thinking particularly of 
what they are saying, but (far'dher, 
lord, stork, drAA*erz) is decidedly 
un-English, and has a Scotch or 
Irish twang with it. See p. 196. 

{31 -i). I use this {u) to represent the 
sound expressed by Mr. M. Bell as 
(soj), see (so). Thus, myrrh, differ 
= {rQ.u,, di\U'). But I do not find 
(i, u) generally distinguished, and 
consequently write (ma, difi*) more 
frequently than (m.^, dif.^*). The 
physiological distinction between 
(go) and (.<) is very difficult to for- 
mulate. See (a), and p. 196. 

(.R .r). This strongly trilled (r) is 
only known as an individual or local 
peculiarity. In Scotland the trilled 
(r) not before vowels, as Jinn (ferm) 
often gives rise to a sensation of (.r), 
as (fe.rm), and many Scots and Irish 
use (.r) as work, arm = (wo.rk, 
ee.rm). It is not recognized ortho- 

(Eh rh) is not now a recognized 
English sound, but is occasionally 
imported from the modern French 
final -re, as sabre (sabrh) for (sabro), 
into the modern English pronuncia- 
tion of anglicised French. Probably 
ags. hr, as it is Icelandic hr. The 
"Welsh rh is rather ('rn) than (rh), 
as generally supposed. 

(S s). From ags. to 19. commonly re- 
presented by s. Eapp imagines the 
ags. sound to have been (sj). In 14. 
(s) was represented s, ss, and by e 
before e, i in words taken from the 
French, and occasionally by sc before 
e, i. In 19. we have the varieties : cell, 
diCe, Glouc(?ster, psdlm., Cirencester 
(S^■s•^■stI), Worcester ("Wwst-j), see, 
scene, coalesce, scAism, Mas/^am, hiss, 
hissed, listen, episde, etc., since 17., 
mispress [vais'iz), sword, breVaska 
(bru-ka), \iel\ows, mezzotint. 


EXPRESSION OF SOUNDS. (Sh— Ui). Chap. VI. $ 2. 

(Sh sh). This was not an ags. sound, 
but it was already developed in 13., 
and it was generally written sch^ but 
sometimes sh, ss, in 13. and 14. 
Orrmin writes sh, ssh, and this was 
used at the end of 15., and generally 
afterwards. At the latter end of 17. 
(sh) was expressed by s before (iu), 
so that siu became (shuu). Traces 
of this found in the early part of 17. 
Towards end of 17. also expressed by 
ci-, si-, sci; ssi; ti-. In 19. we 
have the varieties : cAaise, and fre- 
quently in French words, fucAsia, 
specml, pshviw ! sugar, icAedule, con- 
scioMS,, shaW, wisA^d, Xssheion 
(-iEsh-tun), compresA'/on, mo^«on. 

(T t). From ags. to 19. the regular 
expression is t. In 19., however, 
we have the varieties : de^^, jsicht, 
mdict, sucked, Bouffht, phthisical, re- 
ceipt, toe, thyme, ha^^er, two, mezzo- 

(Th th) was in use from ags. to 19. 
In ags. it was written either ]> or ^, 
or both indifferently. In 13. and 
14. it was sometimes ^, but gene- 
rally ]>, and occasionally th, which 
last expression has remained to 19. 
In 17. in si^A it was written ^h, and 
probably in other words. In 19. we 
have the varieties : Kei^Aley (Kiith*- 
li), eigh^A (f^tth), Ri^ophthegm (sep'o- 
them), SouAampton (SauthHsem*- 
tBu), thin, hlithe (blaith), or (bbidh) 

(Tj tj). An unrecognized English 
sound, generated by the action of a 
following (iu), when the speaker 
avoids the stiffness of (t,j), and 
wishes also to avoid (tsh), as : vir- 
^e, lecture (v.rtju, lek'tjea), com- 
monly (vi'tshu, lek-tsh.i). See (dj). 

(Tsh tsh) was generated, at least, as 
early as 13. from ags. {k), and 
written ch, and in 14. also cch. The 
form ch has remained, but since 16. 
at least cch has become tch, very 
common as a final in 19., in which 
some importations and assimilations 
have produced the varieties : vermz- 
celli, chain, arched, cAioppine, Mar- 
yonbanks (Ma.itsh-b8eqks), match, 

{Tw tw). An unrecognized English 
sound, usually confounded with (tw), 
but it is (t*w) the action of (t) and 
(w) taking place simultaneously, and 
not successively, in twine, twain, etc. 
"Written tw. 

(U u). It is probable that {u) was 

used in 16. at least, and perhaps 
earlier, but it is not easy to dis- 
tinguish (u, u) as short sounds be- 
fore 19., and even then few persons 
acknowledge that pool, pwll, have 
vowels of different quality, as well as 
length (puul, p?d), and that the true 
short sound (u) is heard in French 
poule (pul). Mr. M. Bell considers 
that the Scotch and English pronun- 
ciation of book differ as (buk, bwk). 
To my ears the Scotch have preserved 
also tiae original length of the vowel, 
and say (bunk), or at least give it a 
medial length. Hence, taking (u, 
u) together, we may say that the 
sound has existed and been expressed 
by 11 from ags. to 19. In 14. it was 
also expressed by oii, ow, and the ex- 
pression on was continued in a few 
words in 16., and is not yet quite 
lost as could (k2<d). In 16. (u, u) 
was occasionally expressed by oo, 
still common in ivood, book (wwd, 
louk). In 14. and thence to 16., o 
was often used for (u, u), and is still 
found in a few words. During 17. 
most of the words having (u, u) lost 
the sound, and were pronounced 
generally with (a). There is still a 
fight between {u, a), and in some of 
the Midland Counties the usage is 
just reversed from that now accepted, 
thus (b?it, k«(t, ruh) = but, cut, rub, 
and (fat, pat, fal, bal) = foot, put, 
full, bull. And generally (wad, 
wam'^n) are not uncommon for (wwd 
wum"Bn) =wood, woman. The key 
to this mystery seems to be a pro- 
vincial (a) which becomes labialised 
after labial consonants. In the pro- 
nunciation of the Peak of Derbyshire, 
I have found it very difficult to 
choose between (a, o, tih, u) for such 
words. See below Chap. XI., § 4. 
In 19. we have the varieties: wo- 
man, Bolingbrok(?, wood, worsted, 
Worc^'ster, caoutchouc, could, bwU. 

{Vu). See (u). 

(:U u). This unrecognized English 
sound seems to occur as a variant of 
(y) in Cumberland, Lancashire, and 
East Anglia, and is written as long u. 

(Ui ui), Apparently one of the oldest 
forms of the diphthong oi, oy, pro- 
bably the usual sound in 14., when 
it was also written ui, uy. Still 
used in many words in 16. and even 
17. In the provinces it may be still 
heard in boy (bui), and it is the 
sailor's pronunciation of buoy. 

Chap. VI. § 2. 



(TJu uu). In ags. written m, in 13. w, 
of which this is a characteristic 
orthog-raphy. Between 1280 and 
1310 both u and ou were used. In 
14. ou, ow were generally written, 
but alone was also employed, and 
has remained in many words. In 
16. ow was quite discontinued, and 
ou sparingly used, but oo was intro- 
duced as the usual form, and has 
remained to the present day. How 
soon the (iu) of 17. became (uu) 
after r is not ascertained, but it is 
now the rule (except in the pro- 
vinces), that long u after r = (uu). 
Hence in 19. we have the varieties : 
galleon, Re««ben, '^Vicalengh (Bakluu*), 
hrew, \)xewe&, rheum, rAwbarb, do, 
sho(9, move, mano?Mvre, too, wooed, 
soMp, bo^^se (buuz), ihxough, Broug- 
ham, rendezvous (rondevuu*), surtout 
(sjtuu'), billetdoM^r (bil/duu*), Cotv- 
per,tr^<e, rwling, rule,hruismg,hrmse, 
'H.ulme (Huum), tu'o, who (huu). 

(:I7u uu). A provincial variety of 
(yy), expressed only as long u. 

(V v). In ags. possibly and Orrmin 
(v) was expressed by / between two 
vowels, otherwise it would seem not 
to be an ags. sound. In 13. (v) was 
expressed by u consonant and v con- 
sonant, and so through to 17. when 
V consonant was exclusively applied, 
and u consonant and v vowel discon- 
tinued; but it was seldom repre- 
sented by any but a v form after- 
wards. In 19. we have : of, Belvoir 
(Bii'vi), hsilve, nephew, Grosyenor 
(Groovnj), veal, have, rende:;t^ous. 

(W w). Apparently a peculiar ags. 
sound, and hence expressed by a pe- 
culiar letter p when the Roman 
alphabet of the time was adopted, p. 
613. For this in 13. w was adopted, 
and has remained to 19. The sound 
was sometimes expressed by tt, but 
persuade was often written pcrswade. 
In 19. we have : choir (k^faia:), the 
labial modification assumed as (w), 
see (k?^), pers2^ade, wox. In the 
word one the initial (w), which is 
not written at all, dates probably 
from the latter part of 17. 

(pj[ ui). Defective trill of the lips 
substituted for a trill of the tongue, 
not recognized except as a defect, 
and then written w, but " Lord 
Dundreary" distinguishes (fuiend) 
from (fwend), which last he indig- 
nantly declared he did not say for 

(Wh wh) was probably expressed in 
ags. by hw, and was the ivh of 13. 
to 19. It is still distinctly pro- 
nounced by most northern and careful 
southern speakers, but is rapidly 
disappearing in London. 

(Y y) . This was probably the sound 
of ags. y, and possibly of short u in 
13. It is very doubtful whether 
this short sound has been used at 
all since 13. It seems to have been 
replaced by {i, e). It probably 
occurs, either in this or the cognate 
forms (u, i) in the provinces, and is 
recognized in Scotland. 

{Yy). According to Mr. M. Bell this 
is the indistinct sound only used in 
unaccented syllables in English, and 
written e in : houses, goodness 
(H3uz*yz, gudi'ny&), etc., where or- 
thoepists are doubtful whether it is 
{i) or (e). He also identifies it with 
the "Welsh u, y having a similar 
sound. Not generally recognized, 
and not provided with any distinct 

(Yi yi). The French ui was confused 
with (w«Y) in 16. It is kept in 
some recent words as suite, though 
persons ignorant of French say 
(swiit) . 

(Yy yy) was probably written long y 
in ags. This sound seems to have 
disappeared in 13., or at any rate its 
traces are uncertain. In 14. it re- 
vived with the introduced French 
words, and was written u, eu. It 
remained into 17. written u, eu, ew, 
when it was still recognized by Wal- 
lis, although his contemporary Wil- 
kins seems to have been unable to 
pronounce it, and it was subsequently 
replaced by (iu). It is, however, 
still common in East Anglia, in 
Devonshire, in Lancashire, and pro- 
bably other parts of England, and 
in Scotland, where it appears as a 
substitute for (uu), as was already 
the case in 16. The provincial 
sounds vary as (ii, uu, 99, yy). 

(Z z). Not recognized as distinct 
from (s) in ags. but probably existing 
always, as in 14. it was not unfre- 
quently written z. It has, however, 
been generally confused with s, ex- 
cept in a few words from the Greek. 
The sound seems to have remained 
with few exceptions in the same 
positions from 14. to 19. In 19. 
we have : sacrificing, sacrifice, which 
some pronounce as a substantive with 



Chap. VI. ^ 3. 

(s) and as a verb with (z), <*zar, 
Wim/ior (Wm-za), SaA*bury(SAAlz-- 
bert"), as, discern, e&se, dishonour, 
busmess, sciwors, Ke^i^ick (keztlc), 
he hello It' A', beaux, zeal, sLze, whiz- 
zing, whizzed. 

(Zh zh). Hart 1569 recognized this 
souna in French but not in English. 
Its earliest recognition in English is 
by Miege 1688, who being a French- 
man distinguished it from (sh) with 
which it was long confused. It is 
derived generally from (zi) and hence 
is generally spelled s, z except in 
some recent words, where the Modern 
French sound is employed. In 19. 
we have : rou<7ing, rou<7^, ^eu de 
mots, which "Worcester writes (zhuu*- 
dimoo-) in place of Feline's (zhce d- 
mo), pleasure, divist'on, abscission, 

(') When a mute (p, t, k) ends a 
word, and a pause follows, as the 
contact is loosened, a slight breath 
escapes, not marked in writing, but 

very apparent in (koep', baet*, baekO. 
This was probably always used in 
English, and its absence, which ren- 
ders the consonantdifficultto be heard, 
was probably the occasion of the 
suppression of such final consonants 
in French. 

(') If a sonant (b, d, g) end a word, 
many speakers force out a faint mur- 
muring sound after removing the 
contact, as (eb', xd\ hscg') ebb, add^ 
bagg, similar to the French indica- 
tion of their e miiet in such a place. 
In some speakers this amounts to 
adding (b), and then it is recognized 
in satirical orthography by writing a 
as ebba, adda, bagga. 

{q). The cluck indicated by tut. 

(2,). The cluck indicated by cVch. 

(•) The primary accent Avhich has 
never been indicated in English 

(:) The secondary accent, which has 
never been indicated in English or- 

§ 3. Historical Phonetic Spelling. 

The great multiplicity of forms for the same sound, joined 
to the existing variety of sounds for the same form/ shewn 
in the preceding sections, has urged many persons to attempt 
correcting both by one stroke, as a matter of literature and 
science, and still more with a view to education and uni- 
formity of pronunciation, and with a hope of making our 
language more easy to acquire by foreigners. The device 
has generally consisted either in the introduction of new 
letters, or in giving constant values to known combinations, 
so that the same sound should be always represented by the 
same letters and conversely. In the xii th or xiii th century 
we had Orrmin, in the xvi th Smith, Hart, Bullokar ; in the 
XVII th Gill, Butler, "Wilkins ; in the xviii th, Franklin and 
many others after him in the same and in the xix th century 
both in England and America. The most persistent attempt 
is the phonotypy which grew out of Mr. Isaac Pitman's pho- 
nography or phonetic shorthand, and which in various forms 

^ The strange fantastical variety of 
our orthography, when viewed solely 
from the phonitic point of view, could 
not fail to attract Shakspere's atten- 
tion. Hence he makes Benedick speak 
thus of the love-sick Claudio : " He 
was wont to speake plaine, & to the 

purpose (like an honest man & a souldier) 
and now is he turn'd orthography, his 
words are a very fantasticall banquet, 
iust so many strange dishes." Much 
Ado, ii. 3, speech 5, fo. 1623, p. 107, 
col. 2. 


has been regularly used in printed periodicals from 1843 to 
the present day.^ Such schemes are different from those 
which aim at a universal alphabet for the purposes of science 
or missionary enterprize, such as the alphabets of Max Miiller, 
Lepsius, Merkel, Melville Bell, and the palaeotype used in 
this volume. And neither have the slightest connection with 
the scheme of a universal language, or with any view of 
altering our language in any way, although they have been 
often confounded with such impossibilities. 

After reviewing the two preceding sections the question 
naturally arises : is it possible from the general, firmly estab- 
lished English tcses, to construct a system of orthography which 
should represent our pronunciation at the present day ? If such 
a spelling were possible it would clearly be so suggestive 
that it would be legible to the mere English reader almost 
without instruction. It seems possible, and at least worth 
the trial, for numerous instances occur in which it is ad- 
visable to attempt indicating sounds to purely English 
readers by combinations of the letters with which they are 
familiar. It is also only by exhibiting such a tentative or- 
thography that the possibility of altering our spelling so as 
to more or less indicate our pronunciation, but without alter- 
ing our alphabet, could be properly considered. The follow- 
ing scheme is based upon the two preceding tables, and will 
be termed glossotype, as suggested on p. 13, from its main 
use in compiling provincial glossaries. 

In the phonetic alphabet used by Mr. I. Pitman and myself, 
only 34 simple sounds, 4 vowel diphthongs, and 2 consonant diph- 
thongs, were represented, giving a total of 40 letters in the follow- 
ing order : (ii, ee, aa, aa, oo, uu ; i, e, ae, o, a, w ; oi, oi, 
8U, iu ; J w H ; p b t d tsh dzh kg, f v th dh s z sh zh, r 
1 m n q). The numerous texts which have been printed in this 
alphabet have shown that it suffices for printing our pronunciation 
with sufficient accuracy to satisfy such ears as have not been 
sharpened by a phonetic education. "We may, therefore, commence 
our investigations by determining the best representatives of these 

From the xvi th century ee, oo represent (ii, uu) with certainty, 
from the xviith ai, au represent {ee, aa) with almost, but not 

1 The writer of this treatise was If an alphabet differing entirely from 
much connected with this last scheme the Roman is to be used, and none 
from 1843 to 1849, and in 1848-9 pub- other can be expected to find favour 
lished two editions of the Testament, for all languages, the principles upon 
many books, and a weekly newspaper, which Mr. Melville Bell's various 
the Phonetic News, in the alphabet alphabets of Visible Speech, for print- 
settled by Mr. I. Pitman and himself ing, long and short hand writing, are 
in 1846, which differs in many respects formed, seem to be the best hitherto 
from that now used by Mr. I. Pitman. proposed. 



quite, the same certainty. But there is no usual way of repre- 
senting (oo). The combinations oe, oa are so unfrequent that they 
would occasion hesitation in unusual positions, as : hoepf hoap, 
for hope. Symbols for (aa) have disappeared since the xvii th 
century. The two exclamations oh ! ah ! present the only com- 
binations to which no other value seems to have been assigned ; but 
the combinations oh^ ah, are scarcely used in other words. We 
have then ee, ai, ah, ate, oh, oo, as the only certain represen- 
tatives of the six long vowels (ii, ee, aa, aa, oo, uu). 

The short vowels (i, e) have been uniformly represented by i, e 
from the earliest times, and it would be impossible to obviate the 
ambiguity of their also representing (ai, ii) in accented syllables, 
without pursuing Orrmin's plan and doubling the following con- 
sonant, when it is one of possible initial combination ; thus, vihrait 
would suggest (vai'br^^t), rather than {Yih'veei), which would 
require vihhrait for certainty, and this notation may be adopted 
at the pleasure of the writer. From the xvii th century a, o, u 
have been in like manner the constant representatives of (se, o, a), 
although they would also require duplication of the following 
consonant to preserve them from the ambiguity of {ee, oo, iu), as : 
fammin, notting, fussi = famine, knotting, fussy, compared with : 
famous, noting, fusee =faimus, nohting, Jiwzee, or fyoozee. The last 
short vowel sound (w) occasions gTeat difficulty. In fact it is not 
recognised generally as distinct from (uu), except in such rare 
pairs, as fool full, pool pull. As oo, u have already been appro- 
priated, and as oti, employed for this sound in would, could, should, 
would inevitably suggest the sound (au) in other situations, we are 
driven to some modification of oo, u. The form uh is not English, 
and has been frequently used conventionally for (aa), so that it is 
excluded. The exclamation pooh ! although dictionary makers 
seem only to recognize the orthography pugh, is yet sufficiently 
familiar in the other spelling to all readers,^ and suggests the form 
Qoh for the sound of (w). It is certainly long, but it is known, 
and could only mislead so far as to cause the reader to substitute 
(uu) for {u). The six short vowels are, therefore, i, e, a, o, u, ooh. 
Of the only recognised forms for diphthongs : oy, otv, ew = (oi, 
au, iu), as in hoy, now, new, the first is unobjectionable, but the 
other two do not begin with the elements represented by o, e, (o, e). 
The common diphthong (ai) has no representative distinct from 
i, y, which are already appropriated. Por writing provincial 
dialects a careful separation of the various diphthongal forms is 
important. Hence a systematic mode of representing diphthongs is 
indispensable, and it must be founded upon the historical use of 
y, w, as the second element, which involves the rejection of such 
final forms as ay, aw, for the sounds already symbolised by ai, au. 
By simply prefixing any of the vowels ee, ai, ah, au, oh, oo, 
i, e, a, 0, u, ooh, to y, w, we obtain most suggestive forms 

^ As in Prof. Max Muller's pooh- Lectures on the Science of Language, 
pooh theory of the origin of words, i, 344. 


of diphthongs, containing those vowels run on to a final ee^ oo^ 
typified by the y, w. Thus : aiy {eei) is the usual English -may, — 
ahy (aai), aye, or German ai, — any (AAi), a broad sound of joy, — 
ohy {poi), a provincial sound of boy, — ooy (uui), the Italian \ui, and 
common sailors' 'buoy, — ey (ei), the Scotch bete, — ay (sei), a Cockney 
long i, — oy (oi) the usual boy, — uy (ei) the usual bwy, Gwy ; — eew 
(iiu) an exaggerated Italian iu, — aiw {eeu), an exaggerated Italian 
eu, — ahw (au), the German au, — auw, a broad provincial how, — 
ohw (oou) the common English knoe^ ; — iiv (m) the American and, 
perhaps, the common English new, for which both Wallis and 
Price (p. 139) used the sign iw, — ew (eu) the true Italian eu, — aw 
(aeu) the Norfolk pownd, — ow (ou) a provincial ow, — uw (au) the 
common English now. The use of y, w being only a systematisation 
of an old extinct method of writing diphthongs may be fairly re- 
garded as historical, and gives great power to this system of writing. 

The sounds of (j, w, h) must be represented by y, w, h, having 
no other historic equivalents. But as y, w have been already used 
for diphthongs, and A is a modifying symbol in ah, oh, ooh, in which 
sense it must also be employed amongst the consonant combinations, 
whenever y, w, h occur in such situations as would occasion 
ambiguity, the recognized expedient of inserting a hyphen, as ai-y^ 
oh-w, o-h, must be resorted to. The sound of (wh) must be re- 
presented by the historical symbol wh, instead of the anglosaxon 
hw, which is now uncouth. 

The consonants and consonantal diphthongs must \)Q p I, t d, 
ch j,^ k g, f V, th dh, s z, sh zh, r I m n ng, for although dh, 
zh are unhistorical, they have long been generally recognised as 
orthoepical symbols. To these it seems best to add the historical 
nh for the unhistorical ngk (qk) ; but ngg must be used for (qg) 
to prevent ambiguity, as in singer^ fingger. Hyphens must be 
employed in t-h, d-h, s-h, z-h, n-g, n-k, when each letter represents 
a separate element. All truly doubled consonants must also be 
hyphened, as hoohk-kais, bookcase, distinct from hoohkking, booking, 
and un-ohnd, unowned, from un-nohn, unknown. 

The practical writing alphabet of the English language will 
therefore consist of 42 symbols, which may be fairly called " his- 
torical," namely : ee, ai, ah, au, oh, oo ; i, e, a, o, u, ooh; uy, 
oy, uw, iw ; y, w wh, h ; p h, t d, ch j, k g ; f v, th dh, 
s z, sh zh, r I, m n ng nk. But the use of this alphabet would 
soon point out deficiencies, for example air, ohr, are no adequate 
representatives of the words : air, oar. The indistinct murmur 
which forms the conclusion of these words as generally pronounced 
may be written ('), as the historical representative of an omitted 
found, and the full theoretical sound may be indicated by 'r. This 

1 As these letters are really con- for thth, dhdli, shsh, zhzh (although in 

tractions for tsh dzh, when they are older English ssh is often used for 

doubled to shew that the preceding shsh), because tih represents a really 

vowel is short, it is natural to double different sound, thus Matiln'io would 

only the first element, and write tc]i, dj, be (Mset-thiu) not (Maeth'iu,) and 

meaning ttsh, ddzh. But it is not ai^^A = (Mth), eighth, 
allowable to write tih, ddh, ssh, zzh 




full sound is always heard if another vowel follows, as heeWing^ 
pohWing, pooWeVj Jwfri, /wi^'n'«^ = hearing, pouring, poorer, fiery, 
lowering. Such sounds as her, cur, as distinct from ^^mng, occur' 
rence, require a means of representing the fully trilled r after a 
vowel, as common in Scotland and Ireland, and the examples chosen 
suggests the expedient commonly employed of writing rr, so that 
herd or heWd is English, and herrd is Scotch ' heard.' The voweU 
in '* air, oar, her" however, as distinct from those in ''hale, hole, 
herring," have not yet been represented, and several other signs 
will be found indispensable in writing those dialectic sounds which 
are here of prime importance. 

Now, on examining the long and short vowels, ee i, ai e, ah «, 
au 0, oh u, 00 ooh = (ii i, ee e, aa se, aa o, oo a, uu u), it is readily 
seen that they are more distinct in quality, than in quantity. In 
fact Englishmen find the true short sounds of the long vowels, and 
the true long sounds of the short vowels difficult to distinguish from 
the long and short sounds respectively. This suggests the employ- 
ment of the quantitative signs (~) and ("), when prominence is to be 
given to the quantity, the unmarked sign being regarded as doubt- 
ful, just as in Latin, Italian, Spanish, Welsh, and generally. Thus' 
^en is Scotch, een Yorkshire for the plural of ' eye ' ; tvdit or waigt 
is English, wait Scotch, stohn is Norfolk and American "stone," 
hdok is Scotch, boohk southern English, book northern English, 
'' book," Bath is the local, JBdhth the usual pronunciation of 
*' Bath," and the true sound of ''air" is perhaps e^r, for which ai^r 
is practically sufficient, and the true sound of oar is very nearly, 
but not quite o V. Another way of representing the quantity is the 
thoroughly English method introduced by Orrmin, to which we 
have already found it convenient to have occasional recourse, 
namely, to allow a single following consonant to indicate the length, 
and two following consonants the brevity, of the preceding vowel, 
open vowels remaining ambiguous. Thus the preceding examples 
may be written in order : eenn een, wait waitt, stohnn, hookkj 
hoohlck, look, Bath, Bahth, the short sounds of the two last becoming 
Bathih, Bahthth. Other methods of representing quantity in con- 
nection with accent will be given presently. 

Any one who tried to write down provincial or foreign sounds ■ 
would still find considerable deficiencies. The following sixteen 
additional vowel signs are, however, all that it seems expedient 
to admit, the principle of ambiguous quantity applying as before. 

For ordinary purposes, use : — 
eh={E), for the broader sound of e verging into a, heard in Scotland, 

and generally in the north of England in place of (e), French 

lite, Italian open e. This may also be taken as the sound of 

ai in air, which may be written ehr. 
oa={p), for true sound of oa in o^V = oar, known provincially 

even when not followed by r, a broad sound of oh verging to 

au, Italian open o. 
ui= (y), for Scotch ui, French u, German ii, being ee or rather » j 

pronounced with rounded lips. 


eu={d), for close Prench eu, which has two sounds, close as in 
jeune, and open as mjeune={9, oe), not ordinarily distinguished 
by Englishmen ; the first is ai, the second e or ehy pronounced 
with rounded lips. 

n=(a), to represent French nasality when it occurs, as in enfant ^ 
vin, hon, un, which might be written a'sfa^, on, hoN, un. 

hh, gh=(k\i, gh), for the Scotch and German guttural ch, but (^h) 
may, when desired, be distinguished as yh, and (ke^h, g^h) 
may be written kwh, gwh, 

/^=(lhh) for the common Welch //. 

rr=(.r) for the strongly triUed Scotch r not preceding a vowel, as 

rA=(grh) or (r) for the French, German, and !N'orthumbrian so 
called r grasseye, guttural r or burr. 
For still more accurate dialectic writing, use : 

ae= (ah) for the fine southern ah verging to a. 

aa={aa) for a deeper sound of ah. 

ao={ah) for the broad Scotch ah verging to au. 

uh=(^), for that deeper sound of u which it is necessary to distinguish 
in the provincial diphthongs u?iy, uhw (an, au), if not elsewhere. 

ua={m), for a still deeper sound of w, occasionally heard. 

rio={uh) for the ooh verging to oh, or the oh verging to oo, heard in 
many provincial dialects, the true Italian close o. 

ih, ue={i, u) for the sound of ui verging to ee or oo respectively, as 
heard dialectically in English, German, and French, ih being 
a freqnent form of the German w, and ue being the Swedish u. 

0(?=(oe), for the true German o, and open sound of French eu, de- 
scribed under eu above. 

e or ^=(90), for the sound of it in '' cur," or e in herd, which may be 
written her, herd, (or kor hdrd, if the type e is deficient,) when 
it is considered necessary to distinguish them from Itur, herd. 

d or t? = (jB), for that frequent obscure unaccented a found in canarj, 
real, tenant, which may be written kdnehri, reedl, tenndnt, (or 
if the type d is deficient, hvnehri, reevl, tenuant), when it is 
thought necessary to distinguish it from a or u. 

I or I = (y) for the obscure sound of e goodness, which would be 
written goohdn'iss, (or, if the type i is deficient, goohdmss,) 
when it was thought necessary to distinguish it from e. 

By thus adding from 4 to 12 vowels to the original 12, only 8 un- 
usual, or obscure vowels, out of the 36 recognized in Palaeotype, 
viz., back {oe), mid (y, oh, oh, oh, oh) and front {dh, aeh), are left with- 
out signs, and these probably do not occur in any provincial English 
dialectic pronunciation, but might, in case of necessity, be repre- 
sented by ; ii, eh, uoh, oah, aoh ; euh, oeh, respectively, the first 
two on account of their partial resemblance to the German o, ii, 
and the others on account of their being liable to be confused with 
the sounds already represented by e, uo, oa, ao, eu, oe, respectively. 

The sixteen additional vowel signs are therefore a, aa, ae, ao, e, eh, 
eu, i, ih, oa, oe, ua, ue, uh, ui, uo, and although they are chiefly 



unhistorical, they are so suggestive that they could be readily 
fixed on the memory. Compare aesh a^w^ = ask aunt, in southern 
English, ask ant in fine Yorkshire ; il el English, el ^hl Scotch= 
ill ell; maon Scotch = man, unku geud sheun Scotch = unco guid 
shoon; noa doa^nt goa Norfolk = no don't go ; Goete bdekke German 
= Goethe bocke, muen Devonshire =moon, len Cockney = learn, 
piiir bdhdi Scotch = puir body. 

The system of diphthongs may now be completed by using the 
16 additional vowels as prefixed to y, to ; and also by using all 
the 28 vowels as prefixes to (') and to ui. The (') diphthongs 
are not uncommon provincially, the ui diphthongs are rare, but are 
found in Germany and the Netherlands. The easy method thus 
furnished for representing complicated diphthongal sounds, which 
are so frequently met with in provincial utterances, is one of the 
greatest recommendations for glossotype as a means of writing 
English dialects. 

Any mode of marking the position of the accent is unhistorical, 
but it is so important in unknown words, as all written in Glosso- 
type must be considered, that the Spanish custom of marking its 
position, when not furnished by some simple rule, is well worthy 
of imitation.^ This rule for English has been laid down thus by 
Mr. Melville Bell : The accent is to be read on the first syllable, 
unless otherwise expressed.^ 

The accent mark on an ambiguous vowel or diphthong will be 
the acute on the first portion of the symbol, as reedeem, obtain. The 
accent mark on a short vowel will be the grave, and on a long 
vowel the circumflex, thus combining the notes of quantity and 
accent, as : deemdhnd, deemdhnd^ When the accent falls on more 
than one syllable, it should always be written, as : huywdi^hi^- 
way, Sondhbzdibdlire = unabsehbare, German. The evenness of i 
Erench accent had also best be noted in this way for English 
readers, as di^fdiii = enfant, or otherwise an exception to the rule 
must be made for Erench words only, which would then have to be 
specially named. The small number of accented letters supplied to 
English founts renders it advisable to have a substitute for these 
accent marks, and the turned period used in palaeotype will be 
found most convenient. A device familiar to writers of pronounc- 
ing dictionaries will enable us to indicate the long vowel by placing ' 

* This language seems to be the only 
one, except Greek, in which the ncces • 
sity of marking the position of the ac- 
cent has been acknowledged. In Por- 
tuguese. Italian, English, and Russian, 
the position of 1he accent is a constant 
source of difficulty to foreignei's. The 
Spanish Academy in its anxiety to 
avoid many accent marks, and its desire 
to prevent ambiguity, lays down ^ve 
rather lengthy rules for placing the ac- 
cent mark, which are generally adopted 
by Spanish printers, whether they are 

so by writers I cannot say. "When I 
printed phonetically I carried out a' 
similar system, but the value of it waa 
not sufficiently appreciated, for few or] 
no persons used accents in writing, and] 
Mr. Isaac Pitman, and almost all other \ 
phonetic printers, have utterly ignored] 
accents, at least for all native words. J 
Mr. Melville Bell has however con- 
sistently carried out his one simplij 
rule, which is here recommended to] 
2 Visible Speech for the Million, p. 6. 


the turned period immediately after it, as reesee'd, and tlie short 
vowel by placing it after the following consonant, as empir'ik. 
This principle may be applied to monosyllables, thus readily dis- 
tinguishing : Yorkshire hoo'k, Scotch hooh, English hoohk', with- 
out having to double the following consonant. The principle may 
also be applied to shew the length of the first element of diphthongs, 
so that the true English ^* may know," may be written mdiy nohw, 
or maiy noh'w, while Ictiyd, ndaw or haiyd noaw would indicate 
(b^id nou), which are the Teviotdale pronunciation of "bide, knoll." 

Great care has been bestowed upon this system of writing from 
a belief that it is not a philosophical toy or a plaything, but may 
prove extensively useful to writers of pronouncing vocabularies, to 
provincial glossarists, to travellers forming word lists, to writers of 
Scotch novels, and authors of provincial poems and tales, all of 
whom at present introduce more or less unsystematic, ambiguous, 
or unintelligible orthographies.^ It will be employed, therefore, 
for the representation of dialectic English and Scotch in Chap. XI. 
§ 4. Except for the closest scientific purposes, for which palaeo- 
type, or some system as extensive, is requisite, Glossotype as here 
presented, will be found sufficient.^ 

The practical use of this system of writing^ has suggested some 
improvements in the tabular arrangement, and the preliminary table 
on p. 16, must therefore be considered as cancelled and replaced by 
those on pp. 614-5, In the first of these, the simplest form of 
Glossotype, which may be fairly termed historical phonetic spelling, 
is presented, containing only two of the additional vowels, eu^ ui^ 
without which no dialects could be even approximatively written. 
In the second, these two and the other fourteen are briefly ex- 
plained, some vowel progressions are introduced which may assist 
the reader in forming a conception of the sounds, and the exact 
value of the 28 glossotype vowels, the dijphthongs and consonants 
is fixed by a comparison with palaeotype. 

^ In Mr. Peacock's Glossaries (Tran- to prefix a key conspicuously, but has left 

sactions of the Philological Society, it hidden in a footnote to an appended 

1867, Supplement Part II.) a partially essay, as if it were of no consequence, 

systematic method of writing is adopted, instead of being of prime importance, 

explained in the annexed Essay on One consequence of this to myself was, 

Some Leading Characteristics of the that I did not discover the key till I 

Dialects, etc., p. 11 note ; but on en- had with great difficulty, and much 

deavouring to transliterate the speci- uncertainty, made one for myself by 

mens of the North and South Lonsdale examining the whole glossary. To 

dialects there given (pp. 31, 32) into form a system of writing requires pe- 

glossotype, I found several combina- culiar studies. The present glossotype 

tions and signs employed which had is the result of much thought and ex- 

not been previously explained, and perience extending over a great length 

which I had simply to guess at. Yet of time, combined with long practice 

Mr. Peacock's writing is a gem com- in phonetic writing, 
pared to most which I have met with, ^ Oriental signs can easily be bor- 

for they generally leave me in a state rowed from palaeotype, or supphed by 

of utter bewilderment. Few writers other conventions, 
even condescend to give a key at all, ^ Xhe information from my dialectic 

and in Mr. Peacock's Glossaries, the correspondents (p. 277 note 1) was 

editor has not considered it necessary chiefly collected by means of Glossotype. 



Chap. VI. § 3. 


Especially intended for writing dialectic English according to literary English 
analogies. Isolated letters and words in Glossot)'pe should be in Italics. No letter 
or combination is ever mute ; thus, final e is always pronounced as in German. Never 
use ay, aw, etc., for at, au, etc., even when final. C. Cockney, D. Dutch, £. English, 
F. French, G. German, /. Italian, F. Provincial, S. Scotch, fF. Welsh. 



a gnat 
ah father 
ai wait 
au all 
e net 
ee meet 

i knit 
oh rose 
00 wooed 
ooh wood 
u nut 

eu F, eu ui F. u 

(') an indistinct murmur. 
(j) nasalized utterance. 
N F. nasal n is written n 

Obscure vowels are double 
dotted in her reedl goohd- 
n'is, for which turned letters 
may be used if types run 
short, as : har ree^l goohdnis 

All vowel signs are ambigu- 
ous, short or long, and may 
have their quantity distin- 
guished when desired, by a 
single or double following 
consonant, by the signs of 
quantity (" "), or (^ '), or 
a turned period (•) placed 
immediately after a long 
vowel and after the conso- 
nants following a short 
vowel, as, Yorkshire book 
book book or boo'k, S. bookk 
book book or book', E, 
boohkk boohk boohk or 
boohk', F. noh'w = know, 
Teviotdale noaw =knoll. 

"When accents are not marked 
by (') for ambiguous vowels, 
or C ^ •) for long and short 
vowels as above, the accent 
must be placed in reading 
on the first syllable of a 

ahy G. ai 
aiy mat/ 
ey S. bite 
euy F. ceil 

aw F. C. 
ahw G. au 
aiw C. 
ew /. eu 
euw D. 
iw mew 
ow P. 
ohw know 

uw hoto 

oy log 
uy high 
uiy F. ui 

In all these diphthongs 
the first element has the 
sound assigned in the 
preceding column, which 
is run on quickly, with a 
glide, to a following ee 
or 00 written t/ or w. 
Numerous other diph- 
thongs can be formed on 
the same model. 

Diphthongs may also be 
formed by affixing (') as 
ro/i'(? almost rohad rohud 
— road, and by affixing 
ui, as J). heuuis = huis, 
but it is generally suffi- 
cient to treat this ui as 
y, thus : heuys. 

In the rare cases when any 
of the above combinations 
do not form single vowels 
or diphthongs, introduce 
a hyphen, as ah-y6nt = 
ayont S. Observe that 
the w and y of the conso- 
nants wh, yh, never be- 
long the preceding vowel. 


b hee 
ch chest 
d doe 
dh the 
i fee 

h he 

j >y 

k coo 
kw queen 
kwh 8. quh 
kh G: ach 
1 lo 

Ih W. II 
ly L gl 
m me 
n no 
ng thing 

ngg finger 
nk think 
ny /. gn 
p pea 
r ray 
'r air 
IT I.S. r 
rh F.F. r 
s see 
sh she 
t tin 
th thin 
V vale 
w wail 
wh why 

y yet 

yh S. nicht 
z %eal 
zh vision 

Foreign and Oriental sounds 
must be represented by 
small capitals, &c., by special 

Eeally doubled consonants 
should be separated by a hy- 
phen, as Mw-;2dAw=unknown. 

When any of the above com- 
binations do not form single 
letters introduce a hyphen, 
as mad-huws, Bog-hed, Mak- 
heeth, in-grdin, in-kum, 
mis-hdp, pot-huws, etc. 

Chap. VI. ^ 3. 



Explanation of the Additional and Foeeign Yowels. 

a obscure a in real, cristal. 

aa deeper sound of ah, in G. and F. 

ae between a and ah, fine southern E, a 

in staff, ask, path, pass, command, 
ao between ah and au, broad S. a in man. 
e the obscure sound of e in herd, when it 

can be distinguished from e or u. 
eh between e and a, broad northern E. and 

S. e, I. open e, F. e. 
eu produced by pronouncing at with 

rounded lips, F. close eu in j'eune. 
'i obscure i or e in goodness. 
ih resembling ui verging towards ee, P. G. ii 

oa as heard in oar, between o and oh, 

P. E. broad o, /. open o. 
oe produced by pronouncing e or eh with 

rounded lips, F. open eu in j'eune, G. o. 
ua very deep sound of western E. u. 
ue resembling ui, verging towards oo, 

Swedish u. 
uh deeper and broader sound of u, general 

in P. E. and S. 
produced by pronouncing ee or i with 

rounded lips, 8. ui, D. F. u, G. ii. 
between ok and oo, a broader ooh, 

J. close in somma, Edinburgh coal. 



Vowel Progressions, arranged to shew approximatively how the (italic) sixteen 
additional and foreign vowels lie between the (roman) twelve usual English sounds. 

1. palatal to guttural', ee i ai e eA a ae ah 

2. guttural to labial : ah aa ao au o oa oh uo ooh oo. 

3. labial to palatal : 1) oo ue ui ih ee ; 2) oh oe eu ai 

4. deep to high, obscure : ua uh vl a e i. 

Glossotype Compared with Palaeottpe. 

When more than one palaeotypic symbol is placed after a single vowel, the first 
represents the sound that would be naturally given to it by an English reader, and the 
two may be distinguished, when required, as previously explained. Glossotype in Italics, 
Palaeotype in (). The arrangement is partially systematic. 

Historical, i Additional. 

T series. W series. 

Pairs. , Single. 

ee (ii i) 
ai {ee e) 

ah (aa a) 
au (aa a) 
oh {oo d) 
00 (uu u) 

% {i ii) 
e (e ee) 

a (as 8896) 
(o oo) 
u (o oo) 

ooh {u uu) 

eh (e ee) 
ae (ah aah) 
aa {aa a) i 
ao {ah. adh) 
oa (oo o) 
ue (tju u) 

«^«* (yy y) 

ih (ii i) 
eu {99 9) 
oe (oeoe oe) 
a (^) 

e (aoao oo) 
uh (a: aa:) 
ua (a)) 
uo {uh uuh) 

aiy {eei ei) 
ehy (eI) 
aey (ahi) 
ahy (ai aai) 
aay {ai aai) 
ohy {ooi o\) 
ooy (uui ui) 
uiy (yi) 

ey (ei) 
euy {91 oei) 
ay (aei) 
oy (oi) 
uy (oi) 
uhy (ai) 

aiw {eew. eu) 
ehw (eu) 
aew (ahw) 
ahw (au aau) 
aaw {au) 
ohw {oou ou) 

uiw (yu) 
iw (iu ju) 
ew (eu) 
euw {9u oeu) 
aw (aeu) 
ow (ou) 
uw (ou) 
uhw (au) 

ph{^h) I 
^ 9 (k g) 

ky gy{^ gj) | 

kw gw i^w ' 

wh w (wh. w) 
/i;(fv) I 

th dh (th dli) 
s s (s z) I 

sh %h (sh zh.) I 
chj (tsh dzh) 
yh y (jh hh j) 
Z;A ^A (kh gh) 
X-e<;A y^<^^ (ke^^h 

h (h h') 
r (r) 
V (jr) 
rr (.r) 
Ih (Ihh) 
7 ('1) 

k (Ij) 
w^ (m) 

'm ('m) 

n (n) 

'^ ('n) 

ny (nj) 

w^ (q) 

wZ; (qk) 

Murmur ' (') French Nasals — an ^n on wn (aA ca oa qa). 

The eight omitted palaeotypic vowels may, when required, be indicated by writing — 

ii, eh, uoh, oah, aoh ; euh oeh 
Y, oh, oh oh, oh ; i»h, ddh 

for oe 



The historical spelling from which Glossotype has been evolved, 
is, of course, not proposed for immediate adoption in literature, 
although there is no historical or etymological reason against its 
use. In order to shew the effect of adopting such an orthography 
in place of that now current, I have annexed the glossotypic 
spelling of some lists of words already given in the previous 
section on the pages referred to in each case, in which the reader 
will find the solution of their orthographical riddles. As these lists 
contain the principal anomalies of spelling in our language, the 
absurdity of propagating them will appear strongly in reading over 
their sounds, without having the orthography immediately present 
to the eye. The historical letters only are used, hence the un- 
accented vowels, and some shades of sound are not discriminated 
with perfect accuracy, and the intention has been rather to en- 
deavour to give the letters which an average speller, acquainted 
with the ordinary orthography, would select when intending to 
write his own pronunciation glossotypically, than to aim at or- 
thoepical accuracy, as the appearance which would be presented 
if such a style of spelling were adopted, could not otherwise be 
imitated. For this reason duplicated consonants, are freely ad- 
mitted, when they would be likely to suggest themselves to the 
writer, but are not used systematically, and only the ambiguous 
accent ( ) is employed. The order of the sounds is that given in 
the last paragraph of p. 609. 

ee, p. 599. miniwshiee, deemeen, 
Keez Kolledj, bee, eech, fleed, leev, 
Beechum, leeg, feet, een, kompleet, 
sleev, impreen, Lee, konseet, konseev, 
seenyuri, Lee, reeseet, Beevur, peep'l, 
deemeen, kee, Weemz, keed, duyareea, 
invaleed, greef, maggazeen, greev, 
seenyur, fiwzee, debree, intreeg, feetus, 
kee, keed, muskeetoh, turkeez. 

ai, p. 596. malt, shampain, dailia, 
pain, kampain, strait, trai, haipeni hah- 
peni, jail, Kaishaut'n, gaij, plaig, plai, 
grait, ai ! vail, rain, wai, dhai, ait. 

ah, p. 593. fahdhur, ahr, serahlyoh, 
ah, ahmz, Mahmzben, aikl&,h, ahnt, 
bahrk, klahrk, hahrt, gahrd. 

au, p. 593. faul, aum, Maudlen 
Kolledj, maulstik, wauk, baumun, haul, 
Maud, nauti, Vaun, aun, auful, au, 
braud, sauder, aut, ekstr§,udineri, 
Jaurjik, Jaurj, faurk, baurs. 

oh, p. 602. hohtdr, hohboy, boh, 
yohman, shoh, sohd, pohst, ohts wuts, 
Sohm, bohs'n, kohburn, dob, bobn, 
oblyoh, ob, skrootobr skrootw^ur, yohk 
brobcb, aprohpoh, Grobvnur, deepoh 
deppoh, sobl, robg, Yob-baul, dbob, 
nob, tobrdz, ob, Noblz, kobtb, kwobtb. 

00, p. 605. galoon, Rooben, Bukloo, 
broo, brood, room, roobabrb, doo, shoo. 

moov, manoover, too, wood, soop, booz, 
tbroo. Broom, rondevoo, surtoo, billi- 
doo, Kooper, rooling, troo, rool, brooz- 
ing, brooz, Hoom, too, boo. 

i, p. 599. lanskip, Sinjun, Jahrvis, 
pritti, ginniz, biffin, britcbiz, forfit, 
Tibbuld, huzzif, egzibit, ritb'm, pit, 
marrijiz, marrij, pittid, too liv, siv, 
fippens, wimrain, grits, Jabrvis, Missis, 
bizzi, lettis, bild, biznis, Tirrit, Cbizzik, 
fizzik, Windum, Rotbsi, munni, Anni, 
Biwli, felli, sbammi, plaigi. 

e, p. 595. menni, Pomfret, Pestum, 
Muykel, Temz, sed, Abbergeni, sez, let, 
bed, det, Wenzdi, aledj, forred, beffer, 
Lester, lepperd, cbek, rondevoo, ret- 
turik, frend, konsbens, fettid, konesur, 
berri, ges, pannijerrik, gunnel, Tom- 
masez, saiber, vercboo, Berlingtun, 
saffer, better, Urkert, abuser — or saibur, 
vurcboo, Burlingtun, saflfur, bettur, 
Urkurt, abnsur. 

a, p. 593. sat, Uyzak, Makki, dram, 
bav, banyob, Tammun, plad, sammun, 
barang, Klappam, Talmasb, pik§.nt. 

o, p. 601. rozzin, ounur, on, grot, 
forred, konnisant konnis'nt, Jon, bok, 
peddagog, nolledj. 

w, p. 596. ribbun, meersbum, es- 
kutcbun,umb'l, mobsbun, konsbus, sun, 
duz, luv, tortus, Linkun, find, dub'l, 

Chap. YI. § 3. 



tung, bellus, tuppens, amatur, kubburd, 
avvurdiwpoyz, kurnel, likur, likkur. 

ooh, p. 604. woohmman, Boohlling- 
broohk, woohd, woohstid, Woohstur, 
kuwchoobk, koohd, boobl. 

My, p. 597. nuyv, uyl, duypnossoh- 
fist, huyt, uying, uy, ruynoseros, Ruyn, 
ruyming, ruym, buynd, induyt, duy, 
luyv, suyn, suy, suyd, vuykuwnt^, 
uyl, beeguyling, beeguyl, buy, fluy, 
duy, suydh. 

oy, p. 602. burjoys, noyzi, noyz, 
poynant, koyn, boy, enjoyd, Boyl, koyt 
kwoyt, boy bwoy booy, boyd booyd. 

uw, p. 597. kuwchouk, Makluwd, 
uwr, kuwnter, nuwn, duwt, reenuwns, 
buw, kuw, aluwd. 

iw, p. 599. monniwment, inkiwbait, 
mancbiwmaikar, biwti, fiwd, fiwdal, 
diws, Liwsun, niw, abdiw, viw, viwd, 
fiwg'lmun, amiwzing, fiwsbia, kiw, 
amiwz, kiw, impiwn, biwl, siwt, piwni, 
liwstring, fiwg, iwnuyt, Iwjeen, iw, iw, 
iw, iwl, iw iwtb, or yoo yootb, hiw- 
m&in, hiwman, hiw, Hiw, Hiwz. 

y, p. 600. bidyus, unyun, halilooyah, 
yabrd, Denyil. 

w, p. 605. kwuyr, purswaid, waur, 

wh, p. 605. when. 

h, p. 598. Kala-han, bohl, Koh- 
h6on, bobl. 

p, p. 602. bikkup, pai^ aip, Klap- 
pam, flapper, flapt. 

b, p. 594. bee, eb, ebd, baib, Koh- 
bui-n, Hohburn, kubburd, bohboy. 

t, p. 604. det, yot, induyt, sukt, 
saut, tizzikal, reeseet, tob, tuym, batter, 
too, metsobtint. 

d, p. 594. dellium, deep, ad, Boohd- 
dist, traid, Windum, luvd, woohd, 

ch, p. 604. vairraichelli, chain, 
ahrcht, chopeen, Mahrchbanks, match, 

j, p. 595. Grinnidj,sohljur, judjment, 
ridj, Wedjberi, jem, kolledj, Bellinjam, 

k, p. 600. kan, ak6wnt, Bakkanal, 

Some readers will naturally object to such orthography that it is 
entirely fictitious and not in any respect historical. It is not meant 
to imply that the above spelling was ever used at any time, but 
only that almost every combination of which each word is composed 
has been in use for such a long time, generally more than two centuries, 
that its employment in the sense proposed is really historically justi- 
fied. But how should we spell ? What other grounds of spelling 
are there but the phonetic ? There are the purely historical, the 
etymological, the typographical. The purely historical, however. 

skool, aik, bak, hakt, akwaint, hok, 
kail, baik, wauk, kwak, kee, anteek, 
Urkurt, vuykuwnt, hak'l, eksept. 

g, p. 598. blaggahrd, gob, eg, begd, 
gohst, ges, plaig. 

/, p. 597. fob, fuyf, stif. stuft, fiwg'l- 
man, of n, lahf, hahf, saifer, leftenant. 

V, p. 605. ov, Beevur, hahv, nevviw, 
Grohvnur, veel, hav, rondevoo. 

t]i, p. 604. Keethli, aitth, apohthem, 
Suwth-hamtun, thin, bluyth, bluydh, 

dh, p. 595. dhee, breedh. 

s, p. 603, sel, ais, Gloster Glauster, 
sahm, Sissister, Woohstur, see, seen, 
kohales, siz'm, Massam, hiss, hist, 
lis'n epis'l, missis, sohrd, briska, bellus, 

z, p. 605. sakrifuyzing, sakrifuyz, 
zahr, "Winzur, Saulzberi, az, dizern, 
eez, dizonnur, biznis, sizzerz, Kezzik, 
hee bellohz, bohz, zeel, suyz, whizzing, 

sh, p. 604. shaiz, fiwshia, speshal, 
shau, shoohgger, sheddiwl, konshus, shal, 
wisht, Ashtun, kompreshun, mohshun. 

zh, p. 606. roozhing, roozh, zhoo- 
dimoh, plezhur, divizhun, absizhun, 

r, p. 603. (r), ruyt, retturik, ruyt, 
hurri, katarral, ('r,ii') fee'ring, pai'ring 
debahring, ignoh'ring, poo'rer, fuy'ri, 
buw'ri, (i) spai'r,kaur koh'r,bur, maui'- 
gaij, (.<) mur, deefur, or mer deefer. 

/, p. 600. serahlyoh, maulstik, lais, 
Gilfurd, ail, il, travveld, kil, uyl, brisli, 

m, p. 601. dram, flem, sahm, Chumli, 
am, lam, taim, hammer, shamd, him, 
kammel, Bamf, Pomfret, siz'm, rith'm. 

«, p. 601. stuns'l, ohpning, nau, Jon, 
nob, Kohnbroohk, Kahn, neemonniks, 
kuwntur, kan, ribbun,kain,ippikakkiw- 
§.nna, mannur, Ipand, gunnel, reezning 
niwmattiks, piwni, ohp'n, reez'n. 

ng^ p. 602, fingger, singer, wingd, 
Bermingam, tung, Mingiz — hank, han- 
kerchif, link, drunk, ankshus. 


such as was adopted by the Anglosaxons, and by the best writers in 
the XIII th and xiv th centuiies, was also purely phonetic, reflect- 
ing the pronunciation of the writer to the best of his ability. 
We might adopt that systematised scheme of the xivth century 
explained above (p. 401), and illustrated in the next chapter, 
but we should find it extremely difficult to make any one but an 
Early English student see the value of it, and perhaps even he 
might demur to fixing the time at so recent a period, the latest 
during which the principle of phonetic spelling actually influenced 
the writer. But I know no other period which would in any 
respect answer the purpose. With regard to the words introduced 
since then, we should have to consider how they would have been 
probably pronounced at that time, and write them accordingly. 
The rehabilitation of our orthography on that ground would there- 
fore be a work of extreme difficulty, and would find a correspond- 
ingly small number of adherents. Even those who employed it 
would have to re-memorize every word in the language, a discipline 
to which none would submit who could escape it. The attempt 
to introduce such a system could therefore only result in confusion 
worse confounded. We may adopt it for our xiv th century school- 
books, but we must not ask writers to use it in their everyday 

Dismissing, therefore, any purely historical system, we have 
only to consider the etymological, and the typographical, which 
will occupy the two next sections, while the phonetic ground will 
be considered in the last section. 

§ 4. Etymological Spelling. 

The two tables in §§ 1,2 may serve to dissipate the phantom 
which haunts many brains under the name of etymological ortho- 
graphy. It seems that the gross departure from the original 
phonetic conception which pervades our alphabetic system, and 
which degrades alphabetical to hieroglyphical writing, has led 
persons to suppose that the phonetically useless and inconsistently 
applied letters, which they have constantly to employ, are intended 
to convey to the reader the histoiy and origin of a word, whence it 
came, how it changed, what was its original meaning, and how 
that has been modified. It is true that the recent etymological 
labours of Wedgewood and E. Miiller, might be sufficient to prove 
that such information could not be conveyed by any means, because 
it is in many cases unknown now, and was less known to those 
who have modelled our orthography, and also that when it is 
known, or tolerably certain, there is no generally understood 
abbre\dated system for conveying the information, which often 
requires a considerable amount of words to explain, nor does i 
appear possible to conceive that any such system could be invented, 
much less brought into use. These matters do not strike those 
who are possessed with the etymological conception, for they are 


generally very ill informed respecting the real history of our 
language, and think rather of the recent terms borrowed from the 
Latin and Greek, which present no difficulty whatever, and could 
scarcely be made to present much difficulty by any freak of ortho- 
graphy,^ than of the old terms of Germanic, or N^orman French 
origin, or those, not rare words, in constant use, of which the origin 
is unknown. Many of the troublesome additional letters, which 
were perhaps inserted from a supposed knowledge of the origin of a 
word, are mistakes, few of them are of any assistance, and none of 
them are consistently employed. 

To take a simple example : those who know that oak corre- 
sponds to ags. dc^ may be inclined to think that the h was put in to 
show it was Germanic, and not Latinic or Hellenic, whereas we 
know that the introduction of h was a mere habit of the xrn th and 

XIV th centuries ; or that the inserted a was meant to allude to the 
old «, while the prefixed o shewed the modern change ; whereas, 
we know that the xiv th century wrote simply oh^ ook, that in the 

XV th, and the greater part of the xvrth century, oke was em- 
ployed (this is the orthography of Palsgrave and Levins), and that 
the a was introduced towards the latter end of the xvi th century 
as a mere phonetic contrivance to distinguish {po) from (uu), and 
without any etymological reason whatever. It so happens that we 
still write stroke, nothwithstanding the ags. stracan. There was a 
long fight between sope^ soap, and it is not to be supposed that 
a was carried by Latin sapo. It is but very lately that cloak 
triumphed over clohe ; but there can be no etymological reason, 
because no one is certain of the etymology, and the middle Latin 
clocca, generally adduced, would not favour the a. 

Take another simple instance, which, like the former, applies 
to numerous cases : In the word name, the final e is supposed to 
allude to a former final vowel, and to indicate the lengthening of 
the preceding vowel. The ags. had a final a, but the preceding 
vowel was short. The a had become long in Orrmin's time, and he 
wrote name because he said (naa*me), and not (nam*a), which he 
would have written namma, and similarly he changed all the other 
vowels to accord with his own pronunciation. The meaning of the 
added e was lost in xvth century, and in the xvith it was fre- 
quently, but of course inconsistently, used to indicate vowel length, 
and in this case the length of (aa) as (naam). It was not from 
a wish to preserve the a etymologically that it was not changed to 
naim in the xvii th century, but it was because ai became settled as 
(ee) before name ceased to be (naeaem), so that there was a difi'erence 
in sound felt nearly up to the time when our orthography crystal- 
lized in the xviii th century. Should not we suppose same to give 
us similar information. It would be wrong if it did, for though 
Orrmin has an adjective same, there is no ags. adjective sama, but 
only an ags. adverb same. 

1 Italian : ipoteca, ipotesi, ipofisi, more difficulty than our bishop, and not 
ipofora, filosofo, fisonomia, geroglifico, so much as our church. 
epitaffio, epitalamio, etc., present no 



Chap. VI. $ 4. 

The reason usually given for wishing to retain the ii in spelling 
honour, favour, errour is the French orthography -eur, on the plea 
that this orthography discriminates those words which were taken 
from the French from those where taken direct from the Latin. It is 
certainly not obvious that this discrimination is worth any trouble, 
or that any one could determine to which class every word ending 
in -or or -our really belongs. Nevertheless this etymological reason 
has been frequently advanced, and was especially insisted on by the 
late Archdeacon C, J. Hare.^ Our investigations, however, shew 
that the reason given is altogether fanciful and destitute of any 
foundation of historical truth. These words were spelled -our, in 
the XIV th century, because they were pronounced {-uur), for the 
same reason that jw nu became thou now. Moreover honour could 
not have been derived from honneur, because that French form did 
not exist when the English honour was adopted. The French used 
honor, honur, honour. The mutation of Latin o into French eu did 
not take place till a later period.^ If indeed the French had used 
6U, which they would have pronounced (eu) or (ey), there is no 
doubt that Chaucer who used the sound (eu) and wrote it eu or ew, 
would have also written honeur. We see then that honur has more 
claim than either honor or honour if we go to the old French; 
though honour asserts its right as old English, and just as honos was 
old Latin. But such squabbles are trifling. The historical spelling 
of § 3, would decide in favour of onur or onnur, which no ortho- 
grapher has proposed, although every orthoepist would be scan- 
dalized at the pronunciation of the *' etymological" h. 
" Trouth and honour, fredom and curtesie," 

writes the Harl. MS. 7334, v. 46. "What do we gain, either 
phonetically or etymologically by writing. 

Truth and honor, freedom and courtesy. 
Etymologically, trouth agrees better with ags. treow\e, fredom with 
ags. freodom, curtesie with old French curtesie (Eoquefort).^ The 
spellings true, truth, are certainly etymologically inferior to the 
discarded trewe, trouth, which represented the proper sounds of the 
time, and we ought, on the same principle now, to write troo, 
trooth. The termination -y, used for the threefold termination, -e, 
-ie, -y, the last being a contraction for -iy = «5, is a gross violation 
of all supposed principles of etymological spelling. It is evident 
that those who shaped our spelling had little or no knowledge 
of etymology, had no acquaintance with the customs of our ancient 
orthography, which many even yet regard as a chaos without law, 
or custom, and, except in very rare and very obvious instances, 
paid no attention whatever to historical affiliation, or ancient 

^ On English Orthography, Cam- 
bridge Philological Museum, vol. 1. 

2 Diez, after citing feu jew, heurSy 
pleure, etc., adds " in alien diesen 
Fallen kennt die alte Sprache auch das 

einfache o," Gram, der Rom. Spr. 2nd 
ed. 1856. vol. i. p. 426. 

3 The XIV th century orthography 
of this word is especially considered in 
Chap. VII. § 1, near the beginning. 

Chap. VI. § 5. STANDARD SPELLING. 621 

The first thing which we have to do in studying a new language 
for comparative philology, is to determine its sounds, and only 
in so far as the orthography enables us to determine the sounds, 
is it of any etymological value. Any deviation from phonetic 
representation is an impediment in the way of etymology. And 
the only true etymological spelling which can be conceived is one 
that is strictly phonetic. The investigation which we have just 
concluded, by enabling us to restore from the changing orthography 
the changing sounds, that is, the changing words of our language, 
puts us in a far better position than ever to determine the ety- 
mological relations. We still want a similar investigation for 
Prench, at least, and for all our dialects, as well as that principal 
southern form which alone offered sufficient facilities for examina- 
tion. All the labour and trouble of such an examination would 
have been saved if the writers had had a sufficient alphabet from 
the first, and had known how to use it. Eut, unfortunately, the 
true conditions of alphabetic writing have only just been deter- 
mined, and the number of those who can use correctly even such an 
approximation as is furnished by the forty-two historical phonetic 
symbols of the last section is very small. No one has ever dreamed 
of writing provincial dialects etymologically. It was felt that 
by so doing the whole means of representing them was lost ; for, 
until they were written their etymology could not be determined. 
It was forgotten that our own particular cultivated English lan- 
guage, is but the most fortunate among many dialects, that, 
therefore, its etymology, also, could not be determined till it was 
fixed by phonetic writing, and that, consequently, for etymo- 
logical purposes we should endeavour to represent it on paper 
as accurately as the generality can appreciate it. Other reasons 
there are in abundance. But on the ground of philology alone, we 
can truly say, there is no etymology without phonetics. 

§ 5. On Standard, or Typographical Spelling, 

It is possible to write a language without any relation to 
phonetics. The greater part of the Chinese vocabulary is said to 
be of this nature. One system of writing is prevalent throughout 
a vast empire, is understood by each province, and is provided by 
each with a different set of corresponding vocables. At Pekin they 
cannot understand the speech of Canton, but the writing is 
mutually intelligible. It is like the cyphers of arithmetic, or the 
signs of algebra, and the diagrams of geometry, which are read in 
different tongues, but with the same apprehension of their meaning 
throughout Europe. This ideal has great fascination for many. 
Conceive a grand symboleum, known everywhere, and yet read by 
each in his own tongue. Such a conception has been nearly carried 
out in England, Germany, Erance, and Italy, and probably in other 
countries. A fixed system of spelling has been, either by aca- 



demical authority, or through the action of printers, accepted in 
each country. No two men in England and Germany, at any rate, 
pronounce in the same manner every word wliich they would write 
alike. In Germany completely diverse systems of utterance are 
pursued among the educated in different districts. The high 
German, as distinguished from all and every of these systems, is 
known as *'die Schriftsprache, d. h. als diejenige Sprache in der 
man Deutch schreiht.^^ ^ It is a literary, not a spoken language, and 
in Saxony, in Prussia, on the Ilhine, on the Danube, by the 
Vistula, and the Eider, or in Switzerland, the language changes to 
the ear.'^ The peasantry of Saxony are taught to write High 
German ; their spoken Upper German dialect tries a foreigner sorely. 
In the same way we have a literary language in England, a 
written language, having only a remote connection with the spoken 
tongue, and shaped by printers as an instrument intended to satisfy 
the eye. Indeed the great objection to any innovation is its "odd 
appearance." And persons naturally conceive that to change the 
spelling is to alter the language. We have succeeded in getting 
this orthography to be recognised, and there are probably many 
who look upon it as an institution as unalterable and natural as the 
musical scale (which, by-the-bye, differs materially in different 
countries, and is thoroughly artificial in its origin), and regard any 
unwitting deviation from it as unfitting a person for the commonest 
occupation,^ and excluding him altogether from the ranks of the 
educated, and yet the only ''good (!) spellers" in the country are 
compositors and printers' readers. A reference to the tables in the 
two first sections of this chapter should dissipate all idea of fixed- 
ness, every notion of a sacred character in our orthography. It is 
barely a hundred years old, to give it the longest life. Two 
hundred, three hundred, five hundred years ago our spelling was 
entirely different. The same letters were used, but differently 
collocated, for what only standard orthographers could look upon 
as the same word. ^Notwithstanding this, a standard orthography 
is not only a possibility, but an actuality,* and as long as it is 
accompanied by its indispensable adjunct — a pronouncing dic- 
tionary — it will cease to be detrimental to the philologer, who can 
resort to the phonetic representation for what he requires. But it 
should remain fixed to be of value. However much the language 
may hereafter vary, this crystallized form should remain. No 
change of any kind, or from any cause should be permitted. 

1 " The language of writing, i.e., dialectic pronunciations are mutually 
that language in which we write Ger- unintelligible. 

man," as distinguished from speaking ^ *' Correctness in SpelHng," that is, 

German. K. F. Becker^ Schulgram- habitual use of typographical custom, 

matik der deutschen Sprache, 3rd ed. is essential to those who intend to pass 

1835, § 23. any Civil Service examination. 

2 This is still more striking, I am * The slight variations and uncer- 
informed by natives, in the Arabic tainties pointed out on p. 590, note, 
language. The written symbols and may be entirely disregarded for pre- 
the literary language are the same sent purposes. 

from Morocco to Persia, the native 

Chap. VI. § 5. 



Otherwise to the enormous practical evils of an orthography which 
has no connection with sound, which helps no one to read and no 
one to spell, will be added the last straw of uncertainty. 

For my own part I do not see the value of a standard ortho- 
graphy, but I do see the value of an orthography which reflects 
the pronunciation of the writer. Our present standard orthography 
is simply typographical ; but in that word lies a world of meaning. 
It is a tyrant in possession. It has an army of compositors who 
live by it, an army of pedagogues who teach by it, an army of 
officials who swear by it and denounce any deviation as treason, an 
army, yea a vast host, who having painfully learned it as children, 
cling to it as adults, in dread of having to go through the awful 
process once more, and care not for sacrificing their children to that 
Moloch, through whose fires themselves had to pass, and which 
ignorance makes the countersign of respectability. Accepting this 
fact, I have arranged all my vocabularies according to this typo- 
graphical spelling, simply because it will be familiar to all who 
read this book, and they will, therefore, by its means most readily 
discover what they require.^ But I cannot do so without record- 
ing my own conviction, the result of more than a quarter of a 
century's study, that our present standard typographical spelling is a 
monstrous misshapen changeling, a standing disgrace to our literature. 

^ For the same reason in any dic- 
tionary, whether of ancient or modern 
English, which is published before a 
general revision of our orthography is 
eflfected (the Greek Kalends ?), I re- 
commend an arrangement of the words 
according to the orthography in most 
general use at the time of publication, 
because the intention of such an ar- 
rangement is to find out a word with 
facility, and the most generally used 
orthography is necessarily the one best 
known. No individual systems such 
as "Webster's, or that proposed by Mr. 
E. Jones (p. 590, note), or peculiari- 
ties, such as Mitford's Hand, Milton's 
rime, Johnson's musick, which are not 
found in one book or newspaper in ten 
thousand, should be adopted. Where 
there is a concurrent use, do as Min- 
shew did (supra, p. 104), give all spell- 
ings, the explanations under the one 
thought to be most usual (to the ex- 
clusion of all caprice, individual pre- 
ference, and pet theories of correctness) 
and cross references under the others. 
To search a dictionary of any extent is 
penance enough. The searcher can't 
afford to have his labour increased. 
Would not a beginner in Anglosaxon 
be driven mad by the arrangement in 
Ettmiiller's Dictionary, to which no 

index even is appended ? I have often 
regretted the precious time it has cost 
me. In Dr. Stratman's excellent Dic- 
tionary of the Old English Language 
*' the words are entered in alphabetical 
order, under their oldest form, for ex- 
ample uwen otfgw under a-^en, ^fen, even 
under cefen ; ivel, evel under uvel, etc." 
The consequence is the waste of hours. 
Such a dictionary should have the chief 
article, as in Coleridge's Glossary, un- 
der the most usual existent form, as 
best known, and cross references under 
all the old forms, as being unknown. 
Individual Glossaries must of course 
follow the exact orthography of the 
books which they index, but even here 
cross references may refer to the chief 
article under the usual orthography. 
Great advantage would accrue in com- 
paring all forms of words in all books 
by some such arrangement as this. 
Where the field is so vast and the 
multiplicity of detail so immeasurable, 
those patriotic individuals who give us 
the result of their labours should do 
their best to render them quickly ac- 
cessible. The increased bulk of any 
glossary or dictionary is utterly unim- 
portant, as compared with the saving 
of time to its consulter. 


§ 6. On Standard Pronunciation. 

For at least a century, since Buchanan published his *' Essay 
towards establishing a standard for an elegant and uniform pro- 
nunciation of the English language throughout the British dominions^ 
as practised by the most learned and polite speakers," in 1766, and 
probably for many years previously, there prevailed, and ap- 
parently there still prevails, a belief that it is possible to erect a 
standard of pronunciation which should be acknowledged and 
followed throughout the countries where English is spoken as a 
native tongue, and that in fact that standard already exists, and is 
the norm unconsciously followed by persons who, by rank or educa- 
tion, have most right to establish the custom of speech. 

One after another, for the last century, we have had labourers in 
the field. Buchanan, 1766, was a Scotchman, and his dialect clung 
to him; Sheridan, 1780, was an Irishman, and Johnson, from 
the first, ridiculed the idea of an Irishman teaching Englishmen 
to speak.' Sheridan was an actor, so was Walker, 1791, but the 
latter had the advantage of being an Englishman, and his dic- 
tionary is still in some repute, though those who study it will see his 
vain struggles to reconcile analogy with custom, his constant 
references to the habits of a class of society to which he evidently 
did not belong, his treatment of pronunciation as if determined by 
orthography (precisely in the same way as grammarians consider 
grammar to mould language, whereas both orthography and gram- 

^ " BoswELL : It may be of use, Sir, Lord Chesterfield told me that the 
to have a Dictionary to ascertain the word great should be pronounced to 
pronunciation. Johnson : Why, Sir, rhyme to state ; and Sir William Yonge 
my Dictionary shews you the accent sent me word that it should be pro- 
of words, if you can but remember nouuced so as to rhyme to seaty and 
them. BoswELL : But, Sir, we want that none but Irishmen would pro- 
marks to ascertain the pronunciation nounce it grait. Now here were two 
of the vowels. Sheridan, I believe, men of the highest rank, the one the 
has finished such a work. Johnson : best speaker m the House of Lords, 
Why, Sir, consider how much easier it the other the best speaker in the House 
is to learn a language by the ear, than of Commons, differing entirely." Bos- 
by any marks. Sheridan's Dictionary well's Life of Johnson, anno. 1772, 
may do very well ; but you cannot set. 63. Dr. Johnson, however, had 
always carry it about with you : and, his own fancies : " I perceived that he 
when you want the word, you have not pronounced the word heard, as if spelled 
the Dictionary. It is like the man who with a double e, heerd, instead of 
has a sword that will not draw. It is sounding it herd, as is most usually 
an admirable sword to be sure : but done. He said, his reason was, that if 
while your enemy is cutting your throat it were pronounced herd, there would 
you are unable to use it Besides, Sir, be a single exception from the English 
what entitles Sheridan to fix the pro- pronunciation of the syllable ear, and 
nunciation of English ? He has, in he thought it better not to have that 
the first place, the disadvantage of exception." Ibid, anno 1777, aet. 68. 
being an Irishman : and if he says he Dr. Johnson had forgotten hearty 
will fix it after the example of the best hearken, tvear, bear, to tear, swear^ 
company, why they differ among them- earl, penrl, which all orthoepists of his 
selves. I remember an instance : when time pronounce differently fi'om ear, 
I published the plan for my Dictionary, On great^ seat, see supra, p. 87. 


mar are casts, one of speech sounds, and the other of speech 
combinations); in short, in almost every part of his '' principles," 
and his "remarks" upon particular words throughout his dic- 
tionary, they will see the most evident marks of insufficient 
knowledge, and of that kind of pedantic self-sufficiency which is 
the true growth of half-enlightened ignorance, and may be termed 
" usherism." Walker has done good and hard work; he has laid 
down rules, and hence given definite assertions to be considered, 
and he has undoubtedly materially influenced thousands of people, 
who, more ignorant than himself, looked upon him as an authority. 
But his book has passed away, and his pronunciations are no longer 
accepted. Jones, 1798; Perry, 1805; Enfield, 1807; Fulton, 1821; 
Jameson, 1827 ; Knowles, 1835, need not be more than named. 
The last was a corrector and follower of Sheridan. Smart's Walker 
Remodelled, 1836, and Worcester's Critical and Pronouncing Dic- 
tionary, 1847, are those now most in vogue. Smart was a teacher 
of elocution in London, who enjoyed a considerable reputation; 
Worcester is an American. In both of these we have a distinct 
recognition of the vowels in unaccented syllables, but by no means 
a distinct representation of the same ; and in Smart we have great 
consideration bestowed upon the final vocal r (j), and its dipth- 
thongal action on the preceding vowel. 

The vocabulary of our language is so much more copious than 
the vocabulary of any individual, and the vocabulary of any writer 
is so much more copious than the vocabulary of the same man as a 
speaker — unless he be a public orator, a clergyman, a lecturer, a 
barrister, an actor, — and the orthography of our language conveys 
so little information upon the intended pronunciation of any word, 
that there will be many thousand words that even the most accom- 
plished and varied speakers and hearers have never uttered or heard ; 
and other thousands which they have only on the rarest occasions 
uttered and heard, of the sound of which they must therefore be 
more or less in doubt, unless they feel that confidence in themselves 
which will allow them to assert that their own pronunciation is 
correct, because it is their own.^ By far the greater number of 

^ I do not remember ever meeting words which I never heard pronounced, 

with a person of general education, or From this result some pecuHarities 

even literary habits, who could read off not unworthy of notice. Many of the 

without hesitation, the whole of such a words of my old vocabulary continue to 

list of words as : bourgeois, demy, ac- be pronounced in the provincial dialect 

tinism, velleity, batman, beaufin, bre- in which they were learned, such as tay 

vier, rowlock, fusil, flugleman, vase, for tea, even though I know the right 

tassel, buoy, oboe, archimandrite, etc., pronunciation, and generally recollect 

and give them in each case the same the error after it has been committed, 

pronunciation as is assigned in any I know not that I should regret this, 

given pronouncing dictionary now in as it seems to give to my language a 

use. Dr. Kitto, who lost his hearing living character, which it would neces- 

at twelve years of age, but retained his sarily want, if all framed upon unheard 

power of speech, says : (The Lost models. Many such words do not, 

Senses, 1845, Series 1, Deafness, p. 23) however, occur, as I have exchanged 

"I have often calculated that above many provincialisms for book words, 

two-thirds of my vocabulary consist of which I am not in the same way Hable 



speakers, however, do not feel this confidence, and, afraid that the 
sounds they are accustomed to use in their own limited circles 
would be lidiculcd in the higher walks to which they aspire, are 
glad to take the ''authority" of a pronouncing dictionary as a guide. 
Quis autem custodiet ipsos ctistodes? What guide do the guides follow? 

Now our previous investigation shews that at any given time 
there has always existed a great diversity of pronunciation, and 
that pronunciation has altered with different velocities and in dif- 
ferent directions in different places, that what was considered 
*' polite" at one time, was scouted at another, that there never has 
been so near an approach to a uniform pronunciation as that which 
now prevails, and that that uniformity itself is not Kkely to be so 
great as might have been anticipated. 

Uniformity of pronunciation, necessarily depends upon the 
proximity of speakers. We have seen that the great changes in 
English were produced by the two civil wars, which mixed up 
the elements of our population. In more recent times a certain 
degree of unifoiTaity is sustained, by 1) that communication be- 
tween town and country which disseminates the habits of the 
metropolis throughout the provinces; 2) that system of university 
education which rubs together the different dialects of England 
in a classical mortar, and sends out the product as the utterance of 
young men of rank and fortune, and still more effectively, as that 
of young clergymen throughout the length and breadth of our land, 
and 3) that plan of teaching teachers which instils into them the 
pronunciation of the most usual words and enables them to impress 
it upon their pupils in the primary schools thi'oughout the country. 
But that nothing approaching to real uniformity prevails is easily 
seen, and some striking illustrations will be furnished in Chap. XI. 

When we listen to a discourse we are by natural habit carried 
away with the succession of ideas, and we have great difficulty in 
withdrawing our attention from this, and fixing it merely upon the 
sounds which are uttered. Any one, however, who wishes to study 

to mispronounce. But even my took rections, than from the curious instinct 

words, though, said to be generally pro- which has, in the course of time, been 

nounced with much precision, are liable developed, of avoiding the use of those 

to erroneous utterance through my dis- words about the pronunciation of which 

position to give all such words as they I feel myself uncertain, or which I know 

are written, and it is well known that myself liable to mispronounce. This 

the letters of which many of our words is particularly the case with proper 

are composed, do not adequately re- names and foreign words ; although, 

present the sounds with which they even in such, I am more in dread of 

are pronounced. This error of pro- erroneous quantity than of wrong vo- 

nouncing words as they are written is calization." The above test words, 

the converse of that so common which are not all to be found even in 

among uneducated people, of writing Worcester's dictionary, written in glos- 

words down according to their sounds. sotype according to my pronunciation. 

Many of such faults have, however, would be : burjoys, deemuy, &,ktiniz'm, 

been corrected in the course of years, veleeiti, bauman, bifl&n, breeveer, rul- 

and it may not now be easy to detect luk, iiwzee, fiwg'lman, vauz, tos'l, boy, 

me in many errors of this kind : but ohboy, ahrkim&ndruyt. 
this arises not more from such cor- 



pronunciation must be able to do this. It is entirely insufficient 
and misguiding to ask a person to pronounce you a given word. 
The most you can do is to propound him a sentence, and listen to 
him with closed eyes as he repeats it over and over again. Then 
you will probably detect differences of utterance at each delivery, 
differences which it requires years of care and attention to discrimi- 
nate and symbolize satisfactorily. Even then, too, each delivery 
may be false, that is, not such as the speaker would utter naturally, 
when he was thinking of the meaning and not of the sound of the 
words. Listen to a preacher, shutting out your sense to his mean- 
ing, and observe the alternations of loud, distinct, slow, and scarcely 
audible, obscure, rapid utterances. Listen to the same man en- 
gaged in ordinary conversation, and observe the increase of the 
rapid, obscure utterances, and the difference occasioned in the 
tolerably distinct syllables by the difference of emphasis and de- 
livery. Then think how difficult it is to determine the real pro- 
nunciation of that one man. How much more difficult must it be 
to determine and then bear in mind the pronunciation of thousands 
of other people, whom you only hear occasionally and observe less 
frequently, because you wish to know what, not how, they speak. 
And yet this has to be done by any one who wishes to discover 
what is the real actual existing usage of English speech. It is 
needless to say that it is not done. Certain associations of child- 
hood determine the direction of pronunciation, certain other habits 
and associations of youth and early maturity, serve to modify the 
original, and, if the speaker inclines to consider speech, he may 
artificially '' correct," and at any rate, materially change his habits 
of pronunciation in after life, but this is an exception. He soon 
ceases to hear words, he drinks in ideas, and only glaring differences 
which impede this imbibition, strike him and are, more or less 
falsely, noted. He is in the habit of using an orthography which 
not only does not remind him of the sounds of words, but gives him 
the power of deducing great varieties of pronunciation for unknown 
words. What chance then have we of a uniform pronunciation ? 

"What is the course actually pursued by those who seek to deter- 
mine a standard of pronunciation ? Dr. Johnson laid down as '' the 
best general rule, to consider those as the most elegant speakers 
who deviate least from the written words." ^ This was entirely 
theoretical, and was penned in ignorance of the historical variations 
of the orthoepical significance of the *' written words." "Walker 
asks whether the custom of speech to be followed is the ''usage of 
the multitude of speakers, whether good or bad," epithets which 
beg the question, ''the usage of the studious in schools and colleges, 
with those of the learned professions, or that of those who, from 
their elevated birth or station, give laws to the refinements and 
elegancies of a court ?" and replies that it is " neither of these . . ., 
taken singly, but a sort of compound ratio of all three," which 
expression, knowing what compound ratio means, I do not profess 
to understand. He goes on to say, " JSTeither a finical pronun- 

1 Preface to Dictionary. 


elation of the Court," — (is then Court pronunciation necessarily 
finical?) — ''nor a pedantic Graecism of the schools," — (does this 
eixst?) — "will be denominated respectable usage till a certain 
number," (what propoition ?) "of the general mass of common 
speakers," i.e. those who are neither courtly nor educated? "have 
acknowledged them ; nor will a multitude of common speakers 
authorize," (to whom?) "any pronunciation which is reprobated 
by the learned and polite. As those sounds, therefore," he concludes, 
"which are the most generally received among the learned and 
polite ; as well as the bulk of speakers are the most legitimate," — 
i.e. according to law, but what or whose law ? — " we may conclude 
that a majority of two of these states ought always to concur, 
in order to constitute what is called " by Mr. John Walker, 
" good usage." But how does Mr. John Walker, of Colney 
Hatch, determine the usages of each of the three classes he 
has named, but certainly not defined ? Smart seems to take 
refuge in "the mouth of a well-educated Londoner," presumably 
his own, and he talks of "vulgar speakers," "an appearance of 
pedantry," " quite rustic," "speakers of the old school," "metro- 
politan usage among educated people," "a vulgar mouth," "an 
affected speaker," "the best speakers," " distinct utterance," "ob- 
scure or colloquial utterance," "irrregularity," "vulgarism," 
"current pronunciation," "actual pronunciation," "broad utter- 
ance," "affectation," "the most solemn speaking," "vague and 
fluctuating," "elegant speaker," etc., etc., words and epithets im- 
plying theories or foregone conclusions, but not greatly advancing 
our knowledge. We may then repeat the question, what is the 
course actually pursued by these orthoepical oracles ? It appears 
that they have observed somewhat, thought out, practised and 
taught more, till they have confirmed a usage in themselves, and 
have then announced that usage to be the custom of the "best 
speakers," allowing occasional latitude. W^orcester endeavours to 
judge between past orthoepists, and among them allots the palm to 
Smart, but frequently gives several different pronunciations and 
says that "the reader will feel perfectly authorized" by Mr. Wor- 
cester? "to adopt such a form as he may choose." "The com- 
piler" he adds, "has not intended in any case, to give his own 
sanction" to which, however, he seems to attribute considerable 
weight, " to a form which is not supported by usage," (which he 
has not heard generally used?) " authority," (which some previous 
orthoepist has not recommended?) "or analogy," (as derived from 
orthography?) He most sensibly concludes that "it would be un- 
reasonable for him to make a conformity to his own taste, or to the 
result of his own limited observation, a law to those who may differ 
from him, and yet agree with perhaps the more common usage." 

It has not unfrequently happened that the present writer has 
been appealed to respecting the pronunciation of a word. He 
generally replies that he is accustomed to pronounce it in such 
or such a way, and has often to add that he has heard others 
pronounce it differently, but that he has no means of deciding 


wMcli pronnnciatioii ought to be adopted, or even of saying which 
is the more customary. This, indeed, seems to be the present state 
of the case. A large number of words are pronounced with 
differences very perceptible to those who care to observe, even 
among educated London speakers, meaning those who have gone 
through the usual course of instruction in our superior schools for 
boys and girls. These differences largely increase, if educated 
provincial speakers, especially Scottish, Irish, and Welsh, be taken 
into consideration. If our American brethren are included, the 
diversities still further increase, though our younger colonies 
generally, being of more recent formation, so that few of them can 
count even a small number of persons whose fathers and grand- 
fathers were born and lived in them, do not materially swell the 
number. But if we extend our circle to those who have only 
received primary education, and still more to those who have re- 
ceived no education at all, who, not being able to read and write, 
or having no knowledge of theories of language, have developed 
language organically, we ffud the diversities extremely great. The 
respect which the inferior pays to his superior in rank and wealth 
makes him generally anxious and willing to adopt the pronunciation 
of the superiorly educated, if he can but manage to learn it. 
How can he ? Eeal communication between class and class is all but 
impossible. In London, where there is local proximity, the "upper 
ten," the court and nobles, '* the middle class," the professional, 
the studious, ''the commercial class," the retail tradesman, the 
*' young men and young ladies" employed behind the counter, the 
servants, porters, draymen, artizans, mechanics, skilled and un- 
skilled labourers, market men and women, costermongers, " the 
dangerous classes," — all these are as widely separated as if they 
lived in different countries. Eut almost all read, almost all have 
their favourite periodical, and all such periodicals adopt, within 
narrow limits, the same orthography. If that orthography only 
shewed some kind of pronunciation — it is really of very little im- 
portance which variety of those current among the educated be 
selected, or even if different systems were chosen in different news- 
papers — there would then be some means of comparing pronuncia- 
tions, something less fleeting and more ''questionable" than the 
utterance itself, something to which the reader would in the act of 
reading teach himself to conform. The educated author who has 
fancies of his own respecting pronunciation, could insist on his 
printer " following copy" and giving his opinion in his own spelling. 
But the printers generally, printers of journals in particular, would 
each soon adopt some special form, some vocabulary constructed for 
their office (supra, p. 591, n. 2), and in a few years the jolting of 
these forms together would yield to some compromise which would 
produce the nearest approach to an orthoepical standard we could 
hope to attain. "Would, however, our pronunciation remain fixed ? 
All experience is against its doing so, and consequently spelling 
considered as the mirror of speech, would probably have to be ad- 
justed from generation to generation. 


Is such a standard pronunciation desirable ? The linguist and 
philologist may perhaps sigh over this unnatural and inorganic 
orthopicdic treatment of language. For one, the present writer 
could not suppress a feeling of regret. But the well-being of our 
race points in another direction, llecognizing the extreme import- 
ance of facilitating intercourse between man and man, we should 
feel no doubt, and allow no sentimental regrets to interfere with 
the establishment of something approaching to a general system of 
pronouncing, by means of a general system of indicating our pro- 
nunciation in writing, as far as our own widespread language 
extends. Without in the least presuming to say that other and 
much better systems cannot be devised, the writer may point to the 
historical phonetic spelling, developed in § 3, as a means at hand 
for writing the English language without any new types, with as 
close an adherence to the old orihography, as much ease to old 
readers, and as much correctness in imitating the sounds used by 
the writer at any time, as we could hope to be generally possible. 
And as to primary confusion, what would it matter, if not greater 
than the scarcely observed confusion of speech ? Thus if one writes, 
in this spelling : 

Ahy deemdhnd leev too plahnt maJiy stahf maur furmli on dJiu paJith, 

Wotsiz naiym^ sur? Ahy reeuU dohwnt nohwj mum^ mahy 

memmuriz mizzuruhul : — 
and another writes — 
Ey dimdnd leev took plant mi staf moJtr fermli on dhe path. WJiat 

is hiz naim, ser ? Ey reeali dohnt noh, mam^ mi memmori i% 


both would be intelligible, and a difference of sound not previously 
noticed would be forced on the attention, and probably changed ; 
provided only that those who say ahy plahnt^ &c., will not write ey 
plants etc., because it is "finer," or ''neater," or ** shorter," or 
''nearer to the old orthography," or for any other irrelevant reason, 
which is the great danger to be apprehended — as I know by 

At present there is no standard of pronunciation. There 
are many ways of pronouncing English correctly^ that is 
according to the usage of large numbers of persons of either 
sex in different parts of the country, who have received a 
superior education. All attempts to found a standard of 
pronunciation on our approximate standard of orthography 
are futile. The only chance of attaining to a standard of 
pronunciation is by the introduction of phonetic spelling, 
which will therefore fulfil the conditions required by etymo- 
logical spelling, standard spelling, and standard pronuncia- 
tion. Our present orthography approximately fulfils only 
the second of these conditions, and grossly violates the other 


And thus tlie present writer has been brought round, by a 
totally different route, to the advocacy of a principle to which 
he devoted many years of his life and a considerable portion 
of his means. It is his own conviction, founded not only 
upon philological grounds, but upon philanthropical, educa- 
tional, social, and political considerations, that a phonetic 
system of spelling should be adopted for our noble language. 
To its introduction he finds but one real objection — the exist- 
ence of another orthography. Hitherto all phonetic attempts 
have made shipwreck on this rock. But the enterprising 
spirit of the phonetic navigators is worthy of their arctic 
predecessors, and their aim being not merely to solve a 
problem in natural science, but to increase the power and 
happiness of the vast race which speaks the English language, 
is one which is not likely to die out. Even now a phonetic 
periodical appears regularly in London, conducted by Mr. 
Isaac Pitman, whose widely extended system of phonetic short- 
hand, has done so much to popularize the phonetic idea. 
Even now Mr. Melville Bell has brought out the most philo- 
sophical phonetic alphabet yet invented, and has reduced it 
to a system of writing far simpler and easier than that in 
common use. Even now the present writer is engaged in 
producing a new edition of his Plea for Phonetic Spelling, for 
the second and larger home of our language, the United 
States of America.^ It is true that the difficulties in the 

1 It was in preparing this new edi- missionaries, travellers, etlinologists, 

tion for Mr. Benn Pitman, brother of and philologists ; (7), obscures the real 

Mr. Isaac Pitman, and now of Cincin- history of our language ; (8), conceals 

nati, Ohio, U.S., that I was fortunate the present state of our language; 

enough to discover Salesbury's book (9), hinders the extension and uni- 

(14 Feb., 1859), and thus commenced versal employment of English. Pho- 

the special series of investigations netic Spelling : (1), renders reading 

which have developed into the present very easy ; (2), forms the best intro- 

work. The printing of this third duction to romanic reading ; (3), is 

edition, after the text was complete, as easy as correct speaking; (4), in 

was interrupted by the American Civil conjunction with phonetic reading 

War, and the preparation of these facilitates romanic spelling ; (5), ren- 

pages has hitherto prevented me from ders learning to read even romanically 

finishing the Appendices. It may not a pleasant task; (6), by economising 

be out of place to annex here the head- time, increases the efficiency of primary 

ings from this forthcoming work, pre- schools; (7), affords an excellent logical 

mising that ordinary spelling is therein training to the child's mind; (8), im- 

for convenience termed Romanic. Mo- proves pronunciation and enunciation ; 

manic Spelling : (1), renders reading (9), will greatly assist the missionary 

difficult, and writing still more diffi- traveller and ethnologist; (10), would 

cult; (2), necessitates the memorizing exhibit the real history of our lan- 

of every form in the language; (3), guage ; (11), would exhibit the real 

makes learning to read and write a state of our language; (12), would 

hateful task; (4), is one great cause of induce uniformity of pronunciation; 

our prevailing ignorance; (5), mis-trains (13), would favour the extension and 

a child's mind ; (6), is a hindrance to universal employment of our language ; 



Chap. VI. § 6. 

way are enormous, the dead weight of passive resistance to 
be moved is overwhelming, the ignorance of the active re- 
sisters stupendous, and the hands of the promoters weak ; but 
the cause is good, the direction is historical, the means 
obvious, the end attainable by degrees, the material results 
of even small attempts useful, and one of the most practical 
men that ever spoke or printed our language, Benjamin. 
Franklin, has left on record his own conviction that " some- 
time or other it must be done, or our writing will become 
the same with the Chinese as to the difficulty of learning 
and using it."^ 

(14), would effect a considerable saving 
of printing [this does not apply to 
glossotj-pe, or any system in which 
diagraphs are employed] ; (15), would 
bring phonetic shorthand into general 
use ; (16), would be of material use in 
facilitating etymological investigations. 
The objections considered are arranged 
in five classes; (1). Impossibilities and 
Errors : It is impossible to introduce 
new letters and a new alphabet, or to 
frame a true phonetic alphabet, the 
analysis of all so-called phonetic alpha- 
bets being faulty and insufficient, and 
the new letters hitherto proposed con- 
structed upon an erroneous basis. (2). 
Linguistic Losses : The change from 
romanic to phonetic spelling would 
tend to obscure etjinology, would con- 
fuse words having the same sound but 
different romanic orthography in differ- 
ent senses, would occasion orthography 
to differ from person to person, place 
to place, and time to time, would ob- 
scure idstory and geography, and 

unsettle title deeds by altering the 
appearance of names, and would in- 
troduce vulgarisms of pronunciation. 
(3). Material Losses : The change 
would occasion a great loss of literary 
property, and great expense in pro- 
viding new types. (4). Inconveniences'. 
The change would be bad as change, 
would be too great, and would amount 
to an alteration of the language. (5). 
Difficulties : Phonetic books have a 
strange appearance, we should have to 
learn two systems of spelling instead of 
one, the fewness of the phonetic books 
renders the acquisition of phonetic 
spelling worthless, the change is not 
needed, and is useless, because only 
partially adopted, and another system 
of spelling exists. The author endea- 
vours to shew the incorrectness of all 
these objections, except the last. 

1 The whole of Franklin's remarks 
will be found in a transliteration of 
his own phonetic orthography, infr^ 
Chap. X., § 2. 

































In Part I. 

pp. 270-297. In addition to the arguments there adduced to shew that the 
ancient sound of long i was (n) or (ii), and not (ei, ai, ai), Mr. James A. 
U. Murray has communicated to me some striking proofs from the Gaelic 
forms of English words and names, and English forms of Gaelic names, 
which will he given in Part IV. 

p. 302, 1. 14, blue is erroneously treated as a French word, but in the Alpha- 
betical List on the same page it is correctly given as anglosaxon. The 
corrections which this oversight renders necessary will be given in Part IV., 
in the shape of a cancel for this page, which could not be prepared in time 
for this Part. 

In Fart II. 

p. 442, Faternoster, col. 2, vv. 4 and 8, /or don, miis'doon' read doon, mis'doon*. 
p. 443, Credo 1, col. 2, 11. 4 and 7, /or laverd, ded, read laa-verd, deed; Credo 2, 

col. 2, line 4, /or loverd rend loo'verd. 
p. 462, verses, 1. 2, for Riehard read Richard. 

pp. 464-5. On the use of f for j, and the possibility of } having been occasion- 
ally confused with (s) in speech, Mr. W. W. Skeat calls attention to the 

remarks of Sir F. Madden, in his edition of Lajamon, 3, 437. 
p. 468, Translation, col. 2, 1. 4, /or Ml read hiU. 
p. 473, note, col. 2, 1. 1,/or 446 read 447 ; 1. 17, for (mee, dee, swee, pee) read 

{mee, dee, swee, i^ee) ; 1. 18, for mat/ read May ; 1. 24-5 for (eint'mynt) read 

p. 503, 1, 8, promcnciation, for dead-litshe read dead'liitshe. 
p. 540, 1. 6, for hafSdi read hafSi. 
p. 549, 1. 5 from bottom of text, for mansaugur (maan'sceoei'j^r), read man- 

saungur (maan'soeoeiq-g?r). 
p. 550, Mr. H. Sweet has communicated to me the sounds of Icelandic letters as 

noted by Mr. Melville Bell from the pronunciation of Mr. Hjaltalin, which 

will be given in Part IV. 
p. 553, verse 30, col. 1, 1. 4, for alik&lfii read alikkl^; col. 2, 1. 4, /or aa-li- 

kaaul'vi read aa'likaaul've. 
p. 559, in the Haustlong ; 1. 1,/or er read es, 1. 2, for ex read es ; 1. 4, /or bauge 

read baugi ; 1. 5, for HeMesbror . . . bau'ge 7-ead Hel'ksbror . . . bau'ge ; 

line 7, for isarnleiki read isarnleiki. 
p. 560, note 1, 1. 2, for longr read langr. 
p. 599, col. 2, 1. 14, for demesne read d&mesne. 
p. 600, col. 1, 1. 6, /or Eugene read JSngene. 
p. 614, Glossotype as a system of writing is superseded by Glossic, explained in 

the appendix to the notice prefixed to Part III. 
p. 617, col. 2, under n, 1. 4, /or Ipand read pland. 

In Fart III. 

p. 639, note 2 for (spii's^U", spes-eU') read (spii'sheh", spesh"BU'). 

p. 651. The numbers in the Table on this page are corrected on p. 725. 

p. 653, note 1. The memoir on Pennsylvania German by Prof. S. S. Haldeman, 
was read before the Philological Society on 3 June, 1870, and will be pub- 
lished separately ; Dr. M ombert, having gone to Europe, has not furnished 
any additions to that memoir, which is rich in philological interest. 

p. 680 to p. 725. Some trifling errors in printing the Critical Text and Pronun- 
ciation of Chaucer's Prologue are corrected on p. 724, note. 

p. 754, note 1,/or (abitee'shun) read (abztaa'smn). 

p. 789, col. 1, the reference after famat should be 759*. 

p. 791, col. 2, under much good do it you, for mychyoditio read mychgoditio ; and 
to the references add, p. 938, note 1. 

pp. 919-996. All the references to the Globe Shakspere relate to the issue of 
1864, with which text every one has been verified at press. For later issues, 
the number oii\iQpage (and page only) here given, when it exceeds 1000, 
must be diminished by 3, thus VA 8 (1003), must be read as VA 8 (1000), 
and PT 42 (1057'), must be read as PT 42 (1054'). The cause of this dif- 
ference is that pages 1000, 1001, 1002, in the issue of 1864, containing only 
the single word Poems, have been cancelled in subsequent issues. 


NOTICE, pp. v-xii. 

GLOSSIC, pp. xiii-xx. 

CHAPTER VII. Illustrations of the Pronunciation of English 


§ 1. Chaucer, pp. 633-725. 

Critical Text of Prologue, pp. 633-634. 

Pronunciation of Long U and of AY, EY, as deduced from a com- 
parison of the Orthographies of Seven Manuscripts of the Can- 
terbury Tales, pp. 634-646. 

Treatment of Final E in the Critical Text, pp. 646-648. 

Metrical Peculiarities of Chaucer, pp. 648-649. 

Chaucer's Treatment of French "Words, pp. 650-651. 

Pennsylvania German the Analogue of Chaucer's English, 
pp. 652-663. 

F. W. Gesenius on the Language of Chaucer, pp. 664-671. 

M. Rapp on the Pronunciation of Chaucer, pp. 672-677. 

Instructions for Reading the Phonetic Transcript of the Prologue, 
pp. 677-679. 

Critical Text of the Prologue to the Canterbury Tales, from a 
collation of seven MSS., in a systematic orthography, pp. 680- 
724 (even numbers). 

Conjectured Pronunciation of the same, pp. 681-725 (odd numbers). 
§ 2. Gower, pp. 726-739. 

The Punishment of Nebuchadnezzar, from Gower's " Confessio 
Amantis," Lib. 1, texts of three MSS., and conjectured pronun- 
ciation, pp. 728-737. 

Message from Venus to Chaucer, sent through Gower after his 
Shrift, texts of two MSS., systematic orthography, and con-' 
jectured pronunciation, pp. 738-739. 
§ 3. Wycliffe, pp. 740-742. 

CHAPTER VIII. Illustrations of the Pronunciation of English 


§ 1. William Salesbury's Account of "Welsh Pronunciation, 1567, 

pp. 743-768. 
§ 2. William Salesbury's Account of English Pronunciation, 1547, 

original Welsh text, and translation by Mr. E. Jones, revised by 

Dr. B. Davis, pp. 768-788. 
Index to the English and Latin Words of which the Pronunciation 

is given or indicated in Salesbury's two Tracts, pp. 788-794. 
§ 3. John Hart's Phonetic Writing, 1569, and the Pronunciation of 

French in the xvi th Century, pp. 794-838. 
Account of Hart's original MS., 1551, pp. 794-797, notes. 


Alexander Barcley's French Pronunciation, 1521, pp. 803-814. 
The Lambeth Fragment on French Pronunciation, 1528, 

pp. 814-816. 
Palsgrave on French Pronunciation, 1530, pp. 816-819. 
French Pronunciation according to the French Orthoepists of the 

XVI th Century, pp. 819-835. 
French Orthographic Rules in the xv th Century, pp. 836-838. 
^ 4. William Bullokar's Phonetic Writing, 1580, etc., pp. 838-845. 

English Pronunciation of Latin in the xvi th Century, pp. 843-845. 
§ 5. Alexander Gill's Phonetic Writing, 1621, with an examination of 

Spenser's and Sidney's Rhymes, pp. 845-874. 
Extracts from Spenser's Faerie Queen, with Gill's pronunciation, 

pp. 847-852. 
Extracts from Sir Philip Sidney, Sir John Harrington and other 

poets, with Gill's pronunciation, pp. 852-855. 
Extracts from the Authorized Version of the Psalms, with Gill's 

pronunciation, pp. 855-857. 
An Examination of Spenser's Rhymes, p. 858. 
Faulty Rhymes observed in Moore and Tennyson, pp. 858-862. 
Spenser's Rhymes, pp. 862-871. 
Sir Philip Sidney's Rhymes, pp. 872-874. 
§ 6. Charles Butler's Phonetic Writing, and List of Words Like and 

Unlike, 1633-4, pp. 874-877. 
§ 7. Pronouncing Vocabulary of the xvi th Century, collected from Pals- 
grave 1530, Salesbury 1547, Cheke 1550, Smith 1568, Hart 

1569, Bullokar 1580, Gill 1621, and Butler, 1633, pp. 877-910. 
Extracts from Richard Mulcaster's Elementarie, 1582, pp. 910-915. 
Remarks from an Anonymous Black-letter Book, probably of the 

XVI th Century, pp. 915-917. 
§ 8. On the Pronunciation of Shakspere, pp. 917-996. 
Shakspere's Puns, pp. 920-927. 
Shakspere's Metrical Peculiarities, pp. 927-929. 
Miscellaneous Notes, pp. 929-930. 
Unusual Position of Accents, pp. 930-931. 
Gill on Accent and Metre, pp. 932-939. 
Contracted Words, pp. 939-940. 
Trissyllabic Measures, pp. 940-943. 
Alexandrine Verses, pp. 943-946. 
Shaksperian " Resolutions," Dissyllables corresponding to Modem 

Monosyllables, pp. 947-953. 
Shakspere's Rhymes, pp. 953-966. 

Mr. Richard Grant White's Elizabethan Pronunciation, pp. 966-973. 
Summary of the Conjectured Pronunciation of Shakspere, pp. 973- 

Specimens of the Conjectured Pronunciation of Shakspere, being 

Extracts from his Plays, following the Words of the Folio 

Edition of 1623, with Mod€rn Punctuation and Arrangement, 

pp. 986-996. 


Indisposition, arising from overwork, has greatly delayed the 
appearance of this third part of my work, and a recent relapse, 
rendering the revision of the last seventy pages and the preparation 
of this notice extremely difficult, has compelled me to postpone to 
the next part the illustrations for the xvii th and xviii th centuries, 
which were announced to he included in the present. Three years 
or more will prohably elapse before the remainder of the book can 
be published. 

The fourth and concluding part of this treatise is intended to 
consist of four chapters, two of which, devoted to the xviith and 
xvinth centuries respectively, are now completely ready for press, 
and will therefore certainly appear either under my own or some 
other superintendence. In chapter XI., I am desirous of giving 
some account of Existing Varieties of English Pronunciation, dia- 
lectic, antiquated, American, colonial, and vulgar, for the purpose 
of illustrating the results of the preceding investigation. This can- 
not be properly accomplished without the extensive co-operation of 
persons familiar with each individual dialect and form of speech. I 
invite all those into whose hands these pages may fall to give me 
their assistance, or procure me the assistance of others, in collecting 
materials for this novel and interesting research, which promises to 
be of great philological value, if properly executed. Many hundred 
communications are desirable. There cannot be too many, even 
from the same district, for the purpose of comparison and control. 
As I hope to commence this examination early in 1872, it will be 
an additional favour if the communications are sent as soon as 
possible, and not later than the close of 1871. They should be 
written on small-sized paper, not larger than one of these pages, 
and only on one side, leaving a margin of about an inch at the top 
for reference notes, with the lines wide apart for insertions, and all 
the phonetic part written in characters which cannot be misread. 
Correspondents would much add to the value of their communi- 
cations by giving their full names and addresses, and stating 
the opportunities they have had for collecting the information 
sent. For the purpose of writing all English dialects in one 
alphabet on an English basis, I have improved the Glossotype of 
Chapter VI., and append its new form under the name of Glossic, 
with specimens which will shew the reader how to employ it, 
(pp. xiii-xx.) Eor the sake of uniformity and general intelligibility, 
I should feel obliged if those who favour me with communications 
on this subject would represent all peculiarities of pronunciation 
in the Glossic characters only, without any addition or alteration 
whatever. The little arrangements here suggested will, if carried 



out, save an immense amount of labour in making use of any com- 

The following table will shew the kind of work wanted. All 
the varieties of sound there named are known to exist at present, 
and there are probably many more. It is wished to localize them 
accurately, for the pui-pose of understanding the unmixed dialectic 
English of the xii th and xiii th centuries, and to find traces of the 
pronunciations prevalent in the more mixed forms of the xivth, 
XVI th, and xvii th centuries. Many of the latter will be found in 
Ireland and America, and in the * vulgar' English everywhere. No 
pronunciation should be recorded which has not been actually heard 
from some speaker who uses it naturally and habitually. The older 
peasantry and children who have not been at school preserve the 
dialectic sounds most purely. But the present facilities of com- 
munication are rapidly destroying all traces of our older dialectic 
English. Market women, who attend large towns, have generally 
a mixed style of speech. The daughters of peasants and small 
farmers, on becoming domestic servants, learn a new language, and 
corrupt the genuine Doric of their parents. Peasants do not speak 
naturally to strangers. The ear must also have been long familiar 
with a dialectic utterance to appreciate it thoroughly, and, in order 
to compare that utterance with the Southern, and render it correctly 
into Glossic, long familiarity with the educated London speech is 
also necessary. Resident Clergymen, ^Nonconformist Ministers, 
^National and British Schoolmasters, and Country Gentlemen with 
literary tastes, are in the best position to give the required informa- 
tion, and to these, including all members of the three Societies for 
whom this work has been prepared, I especially appeal. But the 
number of persons more or less interested in our language, who 
have opportunities of observing, is so great, that scarcely any one 
who reads these lines will be unable to furnish at least a few obser- 
vations, and it should be borne in mind that even one or two casual 
remarks lose their isolated character and acquire a new value when 
forwarded for comparison with many others. It is very desirable 
to determine the systems of pronunciation prevalent in the JSTorthem, 
West and East and Central Midland, South Western, South Eastern, 
and purely Eastern dialects. The Salopian, Lincolnshire, and Kent 
Dialects are peculiarly interesting. Mr. James A. H. Murray's 
learned and interesting work on Lowland Scotch (London, Asher, 
1871) will shew what is really wanted for each of our dialectic 

In the following, unfortunately very imperfect. Table a few sug- 
gestive words are added to each combination of letters, and the 
presumed varieties of pronunciation are indicated both in Glossic 
and Palaeotype, but only in reference to the particular combinations 
of letters which head the paragraph. The symbols placed after 
the sign =, shew the various sounds which that combination of 
letters is known to have in some one or other of the exemplificative 
words, in some locality or other where English is the native lan- 
guage of the speaker. In giving information, however, the whole 



word should be written in Glossic, as considerable doubt may 
attacb to local pronunciations of the other letters, and the name of 
the locality, and of the class of speakers, should be annexed. The 
quantity of the vowel and place of the accent should be given in 
every word, according to one of the two systems explained in the 
Key to Universal Glossic, p. xvi, and exhibited on pp. xix and xx. 
In writing single words, the accentual system, used on p. xx, is 
preferable. Great attention should be paid to the analysis of diph- 
thongs, and the Glossic ei, oi, ou, eu, should only be employed where 
the writer, being unable to analyse the sound accurately, confines 
himself to marking vaguely the class to which it belongs. The 
trilled r when occurring without a vowel following should always 
be carefully marked, and the untrilled r should never be marked 
unless it is distinctly heard. Each new word, or item of infor- 
mation, should commence on a new line. Thus : 

cord kaa-d or kdad Eath, workmen, petty traders, etc. 

card ka-d or Md Bath, as before. 

beacon harkn or bdikn Bath, as before. 

key /car or Mi Bath, as before. 

fair feir orfatferfdyerfdyu'' Bath, country farming man. 

Table of Presumed Varieties of English Pronunciation. 

bowels. EA in : leap eat seat meat knead mead 

read speak squeak league leaf leave 
wreathe heath breathe crease ease 
leash weal ear, a tear, seam wean ; 
yea great break bear wear, to tear ; 
leapt sweat instead head thread 
spread heavy heaven weapon leather 
weather measure health wealth = ^e, 
ai, e, ae ; eeh\ aW ; yaa = (ii, ee, 
ee e, ee e; ii', ee', ja.) 

EE in : sheep weed heed seek beef 
beeves teeth seethe fleece trees heel 
seem seen =<?(?, a«; aii/, ey = {n,ee; 
e\, ei) 

EI, EY in : either neither height 
sleight Leigh Leighton conceive 
neive seize convey key prey hey grey 
=ee,^ ai ; aay, uuy, e<y = (ii, ee ; ai, 
ai, 0i). 

EO in : people leopard Leominster 
Leopold Theobald = 6^, e, i, eeoa, 
ecu = {\\, e, i, iioo, iia). 

EU, EW in pew few hew yew ewe 
knew, to mew, the mews, chew Jew 
new shew shrew Shrewsbury stew 
threw sew grew hxew = eeiv, iw, aitt\ 
ew, aew, aw, ui, ue, new, eo, eow, oo, 
oa, oaw uuw ; aa, ah, au ; yoa = (iu, 
m, eu, eu, eu, seu, ii, yy, yu, 99, m, 
uu, 00, 00 w, EU ; aa, a a, a\; joo.) 

I short in : hip crib pit bid sick gig 
stifi", to live, smith smithy withy hiss 
his fish fill swin sin first possible 
charity furniture = f^, i, e, ce, a, u, 
w' = (i, i, e, E, ae, a, b). 

A short in : tap cap bad cat mad sack 
bag ; doubtful in : staflT calf half calve 
halve aftermath path father pass 
cast fast mash wash hand land plant 
ant want hang = «-<?, a, a', aa, ah, an, 
0, ao, o«= (e, ge, ah, a, a, a, 0, 00, 00). 

A long in : gape grape babe gaby late 
skate trade made ache cake ague 
plague safe save swathe bathe pa- 
tience occasion ale pale rare name 
same lane wane=ee, ai, e, ae, a, a\ 
aa; aiy, aih\ ain, ey, eeh' , eeu = {ii, 
ee, ee, ee, seae, aah, aa ; eei, ee', eeQ, 
eei, ii', iig.) 

AI, AY in: way hay pay play bray 
day clay gray say lay may nay, bait 
wait aid maid waif waive ail pail 
trail fair hair chair pair stair = (?^, ai, 
e, ae, aa ; aiy, aay, aa'y = (ii, ee, 
ee, EE, aa; eei, ai, aai.) 

AU, AW in ; paw daw thaw saw law 
raw maw gnaw, bawl maul maunder, 
aunt haunt gaunt daughter =««, ah, 
au, ao, oa ; aaiv, auw = (aa, aa, aa, 
00, 00; au, Au). 

E short in : kept swept neb pretty wet 
wed feckless keg Seth mess guess 
very hell hem hen yes yet = i, e, 
ai, ae, a = [i, e, e, e, ae.) 

E long in : glede complete decent 
extreme here there where me he she 
we \)Q=ee, ai, e, ae, a? =(ii, ee, ee, 
EE, sese ?) 



I long in : wipe gibe kite hide strike 
knife knives wife wives scythe blithe 
ice twice thrice wise pile bile rime 
pine fire shire ; sight right might 
light night fright fight pight ; sight 
rye my lie nigh fry fye pie=i, ee, 
ai, au ; hj, aiy, ey, aay, ahy any, 
tiy, uuy = [a, ii, ee, aa ; a, e\, ei, 
ai, ai, Ai, oi, ai). 

IE in : believe grieve sieve friend fiend 
field yield =<;c, i, e, « ' = (ii, i, i, e, e). 

short, and doubtful, in : mop knob 
knot nod knock fog dog off office 
moth broth brother mother pother 
other moss cross frost pollard Tom 
ton son done gone morning song 
long = o, oa, «o, au, aa, u, uo = {p 00, 

0, 0, A AA, a, 8, u). 

long, A, and OE in : hope rope soap 
note goat oats rodo road oak stroke 
joke rogue oaf loaf loaves oath loth 
loathe goes foes shoes lose roll hold 
gold fold sold home roam hone groan 
= 00, oa, ao, au, ah, aa ; ee, ai ; 
eeh'., ai/i, oah', aoK , oau, aaw, uw, 
uuw ; ye, ya, yaa ; woa = (uu, o oo, 
o 00, AA, aa, aa; ii, ee ; ii', ee\ oo\ 
oo', 000, au, 9u, au, je, jse, ja; woo). 

01, OY in : join loin groin point joint 
joist hoist foist boil oil soil poison 
ointment ; joy hoy toy moil noise 
boisterous foison = oy, any, aay, oay^ 
aoy, uy, uuy, ooy, u ; waay, wuuy, 
ivoy — (oi, Ai, ai, oi, oi, oi, ai, ui, o ; 
wai, wai, woi). 

00 in : hoop hoot soot hood food aloof 
groove sooth soothe ooze tool groom 
room soon moon ; cook look shook 
brook; loose goose =oo, no, tii, ue, 
eo ; eoh\ oeh\ uuw = {m\. u, u, ii, 
yy, 99 ; 99% oe', au). 

OU, OW in: down town now how 
flower sow cow, to bow fleeter e, 
a bow arcus, a bowl of soup 
■cyathus, a bowling green ; plough 
round sound mound hound thou out 
house flour ; found bound ground ; 
our ; brought sought fought bought 
thought ought n(»ught soul four ; 
blow snow below, a low bough, the 
cow lows, a row of barrows, a great 
row tumultus, crow, know ; owe, 
own =00, uo, uo', oa, oa', aa, ah, 
au, ai ; aaw, uw, uuw, oaw, aow, 
uiw, uew,eow, eo^w, oe^w = {\iu.u, uu 
u, uh., 00 0, oh, aa, aa, aa, ee ; au, 
OU, au, oou, oou, lu, yu, ^u, 9j, oey). 

U short in : pup cub but put bud cud 
pudding much judge suck lug sugar 
stuff bluff busy business hush bush 
crush push rush blush bushel cushion 

bull pull hull hulk bulk bury burial 
church rum run punish sung = M, 
uu, uo, oa', i, e, ue, eo — (0, a, u, 
oh, i, e, y, 9). 
U long and Ul, UY in: mute fruit 
bruise cruise, the use, to use, the 
refuse, to refuse, mule true sue fury 
sure union = y 00, eew, ue, uew, ui'w, 
eo, eow, eou = {3uu, in, yy, yu, uu, 

99, 9Uf 9&). 


B mute or —p, f, v, v, w = (p, f, v, 

bh, w). 
C hard and K in : cat card cart sky etc. 

= k, kij, g, gij =^, kj, g, gj), 
C soft = s, sA = (s, sh). 
CH in : beseech church cheese such 

much etc. = cA, h, kh, kyh^ «A = (tsh, 

k, kh, A-h, sh). 
D =^, dh, t, th = {d, dh, t, th). 

G hard in : guard garden, etc. =g, gy , 
y = (g, gj, j), ever heard before n as 
in : gnaw, gnat ? 

G soft, and J in : bridge ridge fidget 
fudge budge =j, g = (dzh, g)_. 

GH in : neigh weigh high thigh nigh 
burgh laugh daughter slaughter 
bough cough hiccough dough chough 
shough though lough clough plough 
furlough, slough of a snake, a deep 
slough, enough through borough 
thorough trough sough tough = mute 
or g, gh, gyh, kh, kyh, f, /', tvh, 
w, 00, p = (g, gh, ^h, kh, kh., f, ph, 
wh, w, u, p). 

H regularly pronounced ? regularly 
mute ? often both, in the wrong 
places ? custom in : honest habita- 
tion humble habit honour exhibi- 
tion prohibition hour hospital host 
hostler hostage hostile shepherd 
cowherd Hebrew hedge herb hermit 
homage Hughes hue humility (h)it 
(h)us ab(h)ominably ? 

J see G soft. 

K see C hard ; ever heard before n in : 
know knit knave knob ? 

L mute in : talk walk balk falcon fault 
vault, alms ? syllabic in : stabl-ing 
juggl-er ? sounded uol, ul, h'l = {ulf 
ol, '1) after o long ? voiceless as Ih ? 

M any varieties ? syllabic in : el-m, 
whel-m, fil-m, wor-m, war-m? 

N nasalizing preceding vowel ? ever = 
iig ? not syllabic in : fall'n, stol'n, 
swoll'n ? 

NG in : long longer hanger danger 
stranger linger finger singer, strength 



length =ng, ngg, nj, n = (q, qg, ndzh, 
n) ; ever ngg or ngk={qg, qk) when 
final in : sing thing nothing ? 

P ever confused with b ? ever post- 
aspirated as p^h = (ph) ? 

QU = kw\ kiv, kwh ?=(kiv, kw, 'kwh. ?). 

E not preceding a vowel ; vocal = r = 
(j), or trilled = r' =(r), or guttural 
= V, *rA = (r, rh), or mute ? How 
does it affect the preceding vowel 
in : far cart wart pert dirt shirt 
short hurt fair care fear shore oar 
court poor ? ever transposed in : 
grass bird etc. ? trilled, and develop- 
ing an additional vowel in : wor-ld 
eur-1 Avor-m wor-k ar-m ? 

E preceding a vowel ; always trilled = 
r' = (r), or guttural = 'r = (r) 
ever labial = ^w, ^br = (m, brh) ? 
Inserted in : draw(r)ing, saw(r)ing, 
law(r) of land, etc. P 

R between vowels : a single trilled r', 
or a vocal /-followed by a trilled r' = 
rr\ h'r' =(ar, 'r) ? 

S =5, z, sh, zli ? = (s, z, sh, zh ?) ; regu- 
larly z ? regularly lisped = ^' A ? = 

SH =^, sA, zh = (s, sh, zh), or, regularly 
2A = (zh)? 

T = ^, ^, th^ s, sh, t Ji = (t, d, th, s, 
sh, tn). 

TH = ^ d, th, tth, dh,f={t, d, th, tth, 
dh, f } in: fifth sixth eighth with 
though whether other nothing etc. 

Y =v,v', w = (bh, w), or regularly lo ? 
W =w, v', v = {\v, bh, v). Is there a 

regular interchange of v, iv ? inserted 
before and 01 in : home hot coat 
point etc. ? regularly omitted in : 
wood wooed would woo wool woman 
womb, etc. ? pronounced at all in : 
write, wring, wrong, wreak, wrought, 
wrap, etc.? any instances of w^ pro- 
nounced as in : lisp wlonk lukewarm 
wlating loathing wlappe wlite ? 

WH =tv, wh, f, f, kwh =(w, wh, f, 
ph, kivh). 

X = k, ks, gz ? 

Y inserted in : ale head, etc. ; regu- 
larly omitted in ye, yield, •ges, yet, 
etc. ? 

7t=z, zh = {z, zh). 

Unaccented Syllables. 

Mark, if possible, the obscure sounds 
which actually replace unaccented 
vowels before and after the accented 
syllable, and especially in the unaccent- 
ed terminations, of which the following 
words are specimens, and in any other 
found noteworthy or peculiar. 

1) -and, husband brigand headland 
midland, 2) -end, dividend legend, 3) 
-ond, diamond almond, 4) -tmd, rubi- 
cund jocund, 5) ~ard, haggard niggard 
sluggard renard leopard, 6) -erd, hal- 
berd shepherd, 7) -ance, guidance de- 
pendance abundance clearance temper- 
ance ignorance resistance, 8) -ence, 
licence confidence dependence patience, 
9) -age, village image manage cabbage 
marriage, 10) -ege, privilege college, 
11) -some, meddlesome irksome quarrel- 
some, 12) -sure, pleasure measure lei- 
sure closure fissure, 13) -tiire, creature 
furniture vulture venture, 14) -ate, [in 
nouns] laureate frigate figurate, 15) al, 
cymbal radical logical cjTiical metrical 
poetical local medial lineal, 16) -el, 
camel pannel apparel, 17) -ol, carol 
wittol, 1 8) -am, madam quondam Clap- 
ham, 19) -om, freedom seldom fathom 
venom, 20) -an, suburban logician his- 
torian Christian metropolitan, and the 
compounds of man, as : woman, etc. , 
21) -en, garden children linen 
woollen, 22) -on, deacon pardon 
fashion legion minion occasion pas- 
sion vocation mention question felon, 
23) -em, eastern cavern, 24) -ar, vicar 
cedar vinegar scholar secular, 25) -er, 
robber chamber member render, 26) 
-or, splendor superior tenor error actor 
victor, 27) -our, labour neighbour 
colour favour, 28) -ant, pendant ser- 
geant infant quadrant assistant truant, 

29) -ent, innocent quiescent president, 

30) -acy, fallacy primacy obstinacy, 31) 
-ancy, infancy tenancy constancy, 32) 
-ency, decency tendency currency, 33) 
-ary, beggary summary granary lite- 
rary notary, 34) -ery, robbery bribery 
gunnery, 35) -ory, priory cursory ora- 
tory victory history, 36) -ury, usury 

Also the terminations separated by a 
h)^hen, in the following words : sof-a 
ide-a, sirr-ah, her-o stucc-o potat-o 
tobacc-o, wid-ow yell-ow fell-ow shad- 
-ow sorr-OAV sparr-ow, val-ue neph-ew 
sher-iff, bann-ock hadd-ock padd-ock 
= frog, poss-ible poss-ibility, stom-ach 
lil-ach, no-tice poul-tiee, prel-acy pol- 
-icy, cer-tain, Lat-in, a sing-ing, a 
be-ing, pulp-it vom-it rabb-it, mouth- 
-ful sorrow-ful, terri-fy signi-fy, child- 
-hood, maiden-head, rap-id viv-id 
tep-id, un-ion commun-ion, par-ish 
per-ish, ol-ive rest-ive, bapt-ize civil- 
-ize, ev-il dev-il, tru-ly sure-ly, har- 
-mony matri-mony, hind -most ut- 
-most better-most fore-most, sweet- 


-ness, right-eous pit-eous plent-eous, 
friend-ship, tire-some whole-some, na- 
-tion na-tiomil, pre-cious prodi-gious, 
ofti-cial par-tial par-tiality, spe-cial 
spc-ciality spe-cialty, ver-dure or-dure, 
fi-gure, in -jure con-jure pcr-jure, plea- 
-sure mea-sure trea-sure lei-sure cock- 
-sure cen-sure pres-sure fis-sure, fea- 
-ture crea-ture minia-ture na-ture 
na-tural litera-ture sta-ture frac-turc 
conjec-ture Icc-ture architcc-ture pic- 
-ture stric-tiire junc-ture punc-ture 
struc-ture cul-ture vul-ture ven-ture 
eap-ture rap-ture scrip-ture depar-ture 
tor-ture pas-tiire ves-ture fu-ture fix- 
-ture sciz-ure, for-ward back-ward 
up-ward down-ward, like-wise side- 
wise, mid-Avif'e house-wife good-wife. 

All inflexional terminations, as in : 
speak-eth speak-sadd-s spok-enpierc-ed 
breath-ed princ-es prince-'s church-es 
church-'s path-s path-'s wolv-es ox-en 
vix-en, etc. Forms of participle and 
verbal noun in -vig. 

Note also the vowel in unaccented 
prefixes, such as those separated by 
a hyphen in the following words : 
a-raong a-stride a-las, ab-use, a-vert, 
ad-vance, ad-apt ad-mire ac-cept af-fix' 
an-nounce ap-pend, a-l-ert', al-cove 
a-byss, auth-entic, be-set be-gin, bin- 
-ocular, con-ceal con-cm- con-trast' 
con-trol, de-pend de-spite de-bate de- 
-stroy de-feat, de-fer', dia-meter, di- 
-rect dis-cuss, e-lope, en-close in-close, 
ex-cept e-vent e-mit ec-lipse, for-bid, 
fore-tell, gain-say, mis-deed mis-guide, 
ob-ject' ob-lige oc-casion op-pose, per- 
-vert, pre-cede pre-fer', pro-mote pro- 
-duce' pro-pose, pur-sue, re-pose, sub- 
-ject' suf-fice, sur-vey sur-pass, sus- 
-pend, to-morrow to-gether, trans-fer 
trans-scribe, un-fit, un-til. 

Position of Accent. 

Mark any words in which unusual, 
peculinr, or variable positions of accent 
have been observed, as : illus'trate 
il'lustrate, demon'strate dem'onstrate, 
ap'plicable applic'able, des'pi cable de- 
spic'able, as'pect aspect', or'deal (two 
syllables) orde'al (three syllables), etc. 


Names of numerals 1, 2, by units to 
20, and by tens to 100, with thousand 
and million. Peculiar names of num- 
bers as : pair, coupte, leash, half dozen, 
dozen, long dozen, gross, long gross, 
half score, score, long score, long hun- 
dred, etc., with interpretation. Pecu- 

liar methods of counting peculiar 
classes of objects. Ordinuls, first, se- 
cond, etc., to twentieth, thirtieth, etc., 
to hundredth, then thousandth and 
millionth. Numeral adverbs : once, 
twice, thrice, four times, some times, 
many times, often, seldom, never, etc., 
Single, simple, double, treble, quadru- 
ple, etc., fourfold, mani-fold, etc., three- 
some, etc. Each, either, neither, both, 
some, several, any, many, enough, enow, 
every. Names of peculiar weights and 
measures or quantities of any kind by 
•which particular kinds of goods are 
bought and sold or hired, with their 
equivalents in imperial weights and 
measures. Names of division of time : 
minute, hour, day, night, week, days 
of week, sevennight, fortnight, month, 
names of months, quarter, half-quarter, 
half, twelvemonth, year, century, age, 
etc., Christmas, Michaelmas, Martin- 
mas, Candlemas, Lammas, Lady Day, 
Midsummer, yule, any special festivals 
or days of settlement. Any Church 
ceremonies, as christening, burying, etc. 

Articles ; the, th', t', e', a, an, etc. 
Demonstratives : this, that, 'at, thick, 
thack, thuck, they = ]7e, them = ]'am, 
thir thor thors these. Personal pro- 
nouns in all cases, especially peculiar 
forms and remnants of old forms, as : 
I me ich 'ch, we us, bus huz, thou thee, 
ye you, he him 'en=hine, shehoo = 
heo her, it hit, its his, they them 
'em = hem, etc. 

Auxiliary verbs : to be, to have, in 
all their forms. Use of shall and will, 
should and would. All irregular or 
peculiar forms of verbs. 

Adverbs and conjunctions : no, yes, 
and, but, yet, how, perhaps, etc. Pre- 
positions : in, to, at, till, from, etc. 

Peculiar syntax and idioms: I are, 
we is, thee loves, thou beest, thou ist, 
he do, they does, I see it = saw it, etc. 

Negative and other contraxjted forms : 
don't doesn't aint aren't ha'nt isn't 
wouldn't couldn't shouldn't musn't 
can't canna won't wunna dinna didn't, 
etc., I'm thou'rt he's we're you're I've 
I'ld Pd I'll, etc. 


The above illustrated in connected 
forms, accented and unaccented, by short 
sentences, introducing the commonest 
verbs : take, do, pray, beg, stand, lie 
down, come, think, find, love, believe, 
shew, stop, sew, sow, must, ought, to 



use, need, lay, please, suffer, live, to 
lead, doubt, eat, drink, taste, mean, 
care, etc., and the nouns and verbs re- 
lating to : bodily parts, food, clothing, 
shelter, family and social relations, 
agriculture and manufacture, processes 
and implements, domestic animals, birds, 
fish, house vermin, heavenly bodies, 
weather, etc. 

Sentences constructed like those of 
French, German, and Teviotdale in 
Glossic, p. xix, to accumulate all the 
peculiarities of dialectic utterances in a 

Every peculiar sentence and word 
should be written fully in Glossic, and 
have its interpretation in ordinary 
language and spelling, as literal as 
possible, and peculiar constructions 
should be explained. 

Comparative Specimen. 

In order to compare different dialects, 
it is advisable to have one passage writ- 
ten in the idiom and pronunciation of 
all. Passages from the Bible are highly 
objectionable. Our next most familiar 
book is, perhaps, Shakspere. The fol- 
lowing extracts from the Ttvo Gentle- 
men of Verona, act 3, sc. 1, sp. 69-133, 
have been selected for their rustic tone, 
several portions having been omitted as 
inappropriate or for brevity. Transla- 
tions into the proper words, idiom, and 
pronunciation of every English dialect 
would be very valuable. 

The Milkmaid, her Virtues and Vices. 

Launce. He lives not now that 
knows me to be in love. Yet I am in 
love. But a team of horse shall not 
pluck that from me, nor who 'tis I 
love — and yet 'tis a woman. But 
what woman, I will not tell myself — 
and yet 'tis a milkmaid. Here is a 
cate-log of her condition. ' Imprimis : 
She can fetch and carry.' "Why a 
horse can do no more ; nay, a horse 
cannot fetch, but only carry ; there- 
fore is she better than a jade. ' Item : 
She can milk ; ' look you, a sweet 
virtue in a maid with clean hands. 

\_Enter Speed. 

Speed. How now ! what news in 
your paper ? 

Launce. The blackest news that 
ever thou heardest. 

Speed. Why, man, how black ? 

Launce. Why, as black as ink. 

Speed. Let me read them. 

Launce. Fie on thee, jolt -head ! 
thou canst not read. 

Speed. Thou liest ; I can. Come, 
fool, come ; try me in thy paper. 

Launce. There ; and Saint Nicholas 
be thy speed ! 

Speed, [^reads'] ' Imprimis : she can 

Launce. Ay, that she can. 

Speed. ' Item : she brews good ale.' 

Launce. And thereof comes the pro- 
verb : ' Blessing of your heart, you 
brew good ale.' 

Speed. ' Item : she can sew.' 

Launce. That's as much as to say. 
Can she so ? 

Speed. ' Item : She can wash and 

Launce. A special virtue ; for then 
she need not be washed and scoured. 

Speed, ' Item : she can spin.' 

Launce. Th'^n may I set the world 
on wheels, when she can spin for her 

Speed. ' Here follow her vices.' 

Launce. Close at the heels of her 

Speed. ' Item : she doth talk in her 

Launce. It's no matter for that, so 
she sleep not in her talk. 

Speed. 'Item : she is slow in words.* 

Launce. villain, that set down 
among her vices ! To be slow in words 
is a woman's only virtue : I pray thee, 
out with't, and place it for her chief 

Speed. ' Item : she is proud.' 

Launce. Out with that too ; it was 
Eve's legacy, and cannot be ta'en from 

Speed. * Item : she will often praise 
her liquor.' 

Launce. If her liquor be good, she 
shall ; if she will not, I will ; for good 
things should be praised. 

Speed. ' Item : she hath more hair 
than wit, and more faults than hairs, 
and more wealth than faults.' 

Launce. Stop there ; I'll have her ; 
she was mine, and not mine, twice or 
thrice in that last article. Rehearse 
that once more. 

Speed. ' Item : She hath more hair 
than wit.' 

Launce. More hair than wit ? It 
may be ; I'll prove it. The cover of 
the salt hides the salt, and therefore it 
is more than the salt : the hair that 
covers the wit is more than the wit, for 
the greater hides the less. What's next ? 


Speed. * And more faults than hairs.' he hath stayed for abetter man than 

Laiince. That's monstrous : 0, that thee, 

that were out ! Speed. And must I go to him ? 

Speed, 'And more wealth than faults.' Launce. Thou must run to hira, for 

Lauuce. Why, that word makes the thou hast stayed so long, that going will 

faults gracious. AVell, I'll have her ; scarce serve the turn, 

and if it be a match, as nothing is im- Speed. Why didst thou not tell me 

possible, — sooner ? pox of your love-letters ! 

Speed. What then ? [Exit. 

Launce. Why, then will I tell thee Launce. Now will he be swinged 

— that thy master stays for thee at the for reading my letter — an unmannerly 

!North-gate. slave, that will thrust himself into 

Speed. For me ? secrets ! I'll after, to rejoice in the 

Launce. For thee ! ay, who art thou ? boys correction. \_Exit. 

Of course it would be impossible to enter upon the subject at 
great leng-th in Chapter XI. The results will have to be given 
almost in a tabular form. But it is highly desirable that a complete 
account of our existing English language should occupy the atten- 
tion of an ENGLISH DIALECT SOCIETY, and I solicit all cor- 
respondents to favour me with their views on this subject, and to 
state whether they would be willing to join such a body. At the 
same time I must request permission, owing to the necessity of 
mental repose on this subject, to abstain from more than simply 
acknowledging the receipt of their communications during 1871. 

In Chap. XII. I hope to consider the various important papers 
which have recently appeared, bearing upon the present investiga- 
tions, especially those by Dr. Weymouth, Mr. Payne, Mr. Murray, 
Mr. Eurnivall, and Herr Ten Brink, together with such criticisms 
on my work as may have appeared before that chapter is printed. 
Any reader who can point out apparent errors and doubtful con- 
clusions, or who can draw my attention to any points requiring 
revision, or supply omissions, or indicate sources of information 
which have been overlooked, will confer a great favour upon me by 
communicating their observations or criticisms within the year 
1871, written in the manner already suggested. The object of 
these considerations, as of my whole work, is, not to establish a 
theory, but to approximate as closely as possible to a recovery of 
Early English Pronunciation. 

Those who have read any portion of my book will feel assured 
that no kind assistance that may thus be given to me will be left 
unacknowledged when published. And as the work is not one for 
private profit, but an entirely gratuitous contribution to the history 
of our language, produced at great cost to the three Societies which 
have honoured me by undertaking its publication, I feel no hesita- 
tion in thus publicly requesting aid to make it more worthy of the 
generosity which has rendered its existence possible. 

Alexander J. Ellis. 

25, Argyll Egad, Kensington, London, W, 
13 February, 1871. 

Appendix to the Notice prefixed to Part III. 




Read the large capital letters alwaijs in the senses they have in the 
following words, which are all i^i the usual spelling except the three 
underlined, meant for foot, then, rouge. 

bEEt bAIt 


cAUl cOAl cOOl 

knIt nEt 


nOt kUt fUOt 



fOTJl eEUd 



WHey Hay 

Pea Bee Toe 


CHest Jest Keep Gape 

EiE Vie THm 


Seal Zeal euSH houZHe 

eaR E'ing eaER'ing 

Lay May Nay siNG 

R is vocal when no vowel follows, and 
modifies the preceding- vowel form- 
ing diphthongs, as in pEER, pAIR, 
bOAR, bOOR, hERb. 

Use R for R' and RR for RR', when 
a vowel follows, except in elemen- 
tary books, where r' is retained. 

Separate th, dh, sh, zh, ng by a 
hyphen (-) when necessary. 

Read a stress on the first syllable 
when not otherwise directed. 

Mark stress by (•) after a long vowel 
or ei, oi, on, eu, and after the first 
consonant following a short vowel. 

Mark emphasis by (*) before a word. 

Pronounce el, em, en, er, ej, a, ob- 
scurely, after the stress syllable. 

When three or more letters come to- 
gether of which the two Jirst may 
form a digraph, read them as such. 

Letters retain their usual names, and 
alphabetical arrangement. 

Words in customary or NOMIC spell- 
ing occurring among GLOSSIC, 
and conversely, should be underlined 
with a wavy line ^-^^^^^ and printed 
with spaist letters, or else in 
a different type. 

Spesimen ov Ingglish Glosik. 

NoM'iK, (clhat iz, kustemeri Ingglish speling, soa kauld from 
dhi Greek nonvos, kustem,) konvai'z noa intimarslien ov dhi 
risee'vd proanunsiai'shen ov eni werd. It iz konsikwentli veri 
difikelt too lem too reed, and stil moar difikelt too lern too reit. 

Ingglish Glosik (soa kauld from dhi Greek gloas'sa, tung) 
konvai-z whotever proanunsiai'shen iz inten'ded bei dhi reiter. 
Glosik buoks kan dhairfoar bee maid too impaar't risee'vd 
aurthoa'ipi too aul reederz. 

Ingglish Glosik iz veri eezi too reed. Widh proper training, a 
cheild ov foar yeerz oald kan bee redili taut too giv dhi egzak't 
sound ov eni glosik werd prizen'ted too him. Aafter hee haz 
akwei'rd familiar'iti widh glosik reeding hee kan lern nomik 
reeding aulmoast widhou't instruk'shen. Dhi hoal teim rikwei'rd 
faur leming hoath glosik and nomik, iz not haaf dhat rikwei'rd 
faur lerning nomik aloa'n. Dhis iz impoa'rtent, az nomik buoks 
and paiperz aar dhi oanli egzis'ting soarsez ov infermai'shen. 



Glosik reiting iz akweiTd in did proases ov glosik reeding. Eni 
wun hoo kan reed glosik, kan reit eni werd az wel az hee kan 
speek it, and dhi proper moad ov speeking iz lemt bei reeding 
glosik buoks. But oaing too its pikeu'lier konstiiik'sben, glosik 
speling iz imee"dietli intel'ijibl, widhou't a kee, too eni nomik 
reeder. Hens, a glosik reiter kan komeu-nikait widh mil reederz, 
wliedher glosik aur nomik, and haz dbairfoar noa need too bikum- 
a nomik reiter. But hee "kan bikum* wun, if serkemstensez render 
it dizei'rrabl, widh les trubl dhan dhoaz hoo hav not lemt glosik. 

Dhi novelti ov dhi prezent skeem faur deeling widh dhi Speling 
Difikelti iz, that, wheil it maiks noa chainj in dhi habits ov egzis'- 
ting reederz and reiterz, and graitli fasiMtaits leming too reed our 
prezent buoks, it entei'rli obviaits dhi nisesiti ov leming too reit 
in dhi euzheuel komplikaited fashen. 

Dhi abuv* aar edeukai'shenel and soashel eusez ov Glosic. It 
iz heer introadeu'st soalli az a mecnz ov reiting Aul Egzisting 
Varei'itiz ov Ingglish Proanunsiai'shen ^ bei meenz ov Wun Alfa- 
bet on a wel noan Ingglish baisis. 

^ Eevn araung* heili edcukaited Ing- 
glishraen, maarkt varei'itis ov proa- 
nunsiai'sken egzis-t. If wee inldood 
proavin-shel deialekts and vulgar'itiz, 
dhi number ov dlieez varei'itiz wil bee 
inau-rmusli inkrecst. Dhi eer ri- 
kwei-rz much training, bifoar it iz 
aibl too apreeshiait mineu't shaidz ov 
sound, dhoa it redili diskrim-inaits 
brand diferensez. Too meet dhis difi- 
kelti dhis skeem haz been diveided mtoo 
•too. Dhi ferst, aur Ingglish Glosik, 
iz adap'ted faur reiting Ingglish az wel 
az dhi autherz ov proanounsing dik- 
sheneriz euzheueli kontemplait. Dhi 
sekend aur Euniversel Glosik, aimz at 
giving simbelz faur dhi moast mineu't 
foanet-ik anal'isis yet achee'vd. Dhus, 
in dhi ferst, dhi foar difthongz ei, oi, 
ou, eu, aar striktli konven'shenel seinz, 
and pai noa heed too dhi grait varei'iti 
ov waiz in which at leest sura ov dhem 
aar habit' eueli proanou'nst. Agai*n, 
eer, air, oa'^, oor, aar stil ritn widh ee, 
ai, oa, 00, auldhoa- an aten-tiv lisner 
wil redili rekogneiz a mineut aulte- 
rai-shen in dheir soundz. Too fasikitait 
reiting wee mai euz el, em, en, ej, a, 
when not under dhi stres, faur dhoaz 
obskeu'r soundz which aar soa preva- 
lent in speech, dhoa reprobaited bei 
aurthoa-ipists, and singk dhi disting-k- 
shen bitwee-n i, and ee, under dhi saim 
serkemstensez. Aulsoa dhi sounds in 
defer, occur, deferring, occur- 
ring may bee aulwaiz ritn with er, 
dhus difer', oker', diferring, oJcer'ring, 
dhi dublino: ov dhi r in dhi 'too laast 

werdz sikeu-rring dhi voakel karakter 
ov dhi ferst r, and dhi tril ov dhi 
sekend, and dhus disting'gwishing 
dheez soundz from dhoaz herd in her'- 
ing^ okuvens. Konsid*erabl ekspee'r- 
riens sujes'ts dhiz az a konvee-nient 
praktikel aurthoa-ipi. But faur dhi 
reprizentai'shen ov deialekts, wee re- 
kwei'r jenereli a much strikter noatai*- 
shen, and faur aurthoaep'ikel diskrip'- 
shen, aur seientifik foanet'ik dis- 
kush'en, sumthing stil moar painfuoli 
mineu't. A feu sentensez aar anek'st, 
az dhai aar renderd bei Wauker and 
Melvil Bel, ading dhi Autherz oan 
koloa-kwiel uterens, az wel az hee kan 
estimait it. 

Praktikel. Endever faur dhi best, 
and proavei-d agen-st dhi werst. Ni- 
sesiti iz dhi mudher ov inven'shen. 
Hee* hoo wonts konten*t kanot feind 
an eezi chair. 

Wauker. Endevur faur dhe best, 
and pr'oavaayd agen*st dhe wurst. 
Neeses-eetee iz dhe mudh-ur ov inven*- 
shun. Hee* hoo wonts konten*t kan*- 
not faaynd an eezee chai-r. 

Melvil Bel. Endaevu'r fo'r dhi' 
baest, a'nd pr'aovaayd a'gaenh'st dhi' 
wuurst. Neesaes'iti iz dhi' muudhu'r 
o'v invaenh-shu'n. Hee* hoo waunh'ts 
ko'ntaenh't kano't faaynd a'n ee'zi 

Elis. Endevu' fu')dhi)bes-t u'n)- 
pr'oa'vuyd u'gen'st dhi)wu-st. Ni- 
ses-iti)z dhi)mudh'u'r' u'v)inven-shu'n. 
Hee- hoo)won'ts ku'nten*t kan'ut fuynd 
u'n)eezi che"u'. 




Small Capitals throughout indicate 
English Glossic Characters as on p. xiii. 
Large capitals point out the most im- 
portant additional vowel signs. 

The Thirty-six Vowels of Mr. A. 
Melville Bell's "Visible Speech." 








uu' ea ee 

U' I' I 



AA A' E 


ua ua' AE 

AH E' A 


Wide Round. 


GO ui' ui 

uo uo' UE 


OA oa' EO. 

AO ao' OE 


Au au' eo' 

o' oe' 

Brief Key to the Vowels. 

A as in English gnat. 

A' (read ai-hiiok) fine southern Eng- 
lish asJc, between aa and e. 

AA as in English baa. 

AE usual provincial English e^ French 
e, German a. 

AH broad German ah, between aa & au. 

AI as in English bait, with no after- 
sound of ee. 

AO open Italian o, between o and oa. 

ao' closer sound of ao, not quite oa. 

AU as in English caul. 

au closer sound of au, as iin Irish sir. 

E as in southern English net. 

E' modification of e by vocal r in herb. 

ea Russian ti, Polish g, variety of ee. 

EE as in English beet. 

EO close French eu in peu, feu. 

eo' opener sound of eo, not quite oe. 

I as in English knit. 

V opener sound of i, not quite e, 
as e in English houses, Welsh n. 

o as in English not, opener than au. 

o' a closer sound of o. 

OA as in English coal, with no after- 
sound of 00. 

oa' closer sound of oa; u with lips 

OE open French eu in veuf, German o, 

oe' opener sound of oe, 

00 as in English cool. 

u as in English nut. 

U' obscure u, as o in English mention. 

ua open provincial variety of u. 

ua' slightly closer ua. 

UE French m, German ii. 

ui provincial Ger. ii, nearly ee^ Swed. y. 

ui' Swedish long u. 

uo as in English full, woman, book, 
uo' Swedish long o. 
UU usual provincial variety of u. 
uu' Gaelic sound of ao in laogh ; try 
to pronounce oo with open lips. 

Special Rules for Vowels. 

Ascertain carefully the received pro- 
nunciation of the first 12 key words on 
p. xiii, (avoiding the after-sounds of ee 
and 00, very commonly perceptible after 
ai and oa) . Observe that the tip of the 
tongue is depressed and the middle or 
front of the tongue raised for all of 
them, except u ; and that the lips are 
more or less rounded for oo, uo, oa, 
au, 0. Observe that for i, e, uo, the 
parts of the mouth and throat be- 
hind the narrowest passage between 
the tongue and palate, are more widely 
opened than for ee, ai, oo. 

Having ee quite clear and distinct, 
like the Italian, Spanish, French, and 
German i long, practise it before all 
the English consonants, making it as 
long and as short as possible, and when 
short remark the difi'erence between 
ee and i, the French f,ni, and English 
finny. Then lengthen i, noticing the 
distinction between leap lip, steal still, 
feet ft, when the latter words are sung 
to a long note. Sustaining the sound 
first of ee and then of i, bring the lips 
together and open them alternately, 
observing the new sounds generated, 
which will be ui and ue. A proper 
appreciation of the vowels, primary ee, 
wide i, round ui, wide round ue, will 
render all the others easy. 

Obtain oo quite clear and distinct, 
like Italian and German u long, French 
ou long. Pronounce it long and short 
before all the English consonants. Ob- 
serve the distinction between pool and 
pull, the former having oo, the latter uo. 
The true short oo is heard in French 
poule. English pull and French poule, 
differ as English fnny and French 
fni, by widening. Observe that the 
back of the tongue is decidedly raised 
as near to the soft palate for oo, uo, as 
the front was to the hard palate for 
ee, i ; and that the lips are rounded. 
While continuing to pronounce oo or 
uo, open the lips without moving the 
tongue. This will be difiicult to do 
voluntarily at first, and the lips should 
be mechanically opened by the fingers 
till the habit is obtained. The results 
are the peculiar indistinct sounds uu 



and u\ of which w' is one of our com- 
monest obscure and unaccented sounds. 

In uttering ee, ai, ai; the narrowing 
of the passage between the tongue and 
hard pahite is made by the middle or 
front of the tongue, which is gradually 
more retracted. The ai, ae, are the 
French e, e, Italian e chiuso and 
e aperto. The last ae is very common, 
when short, in many English mouths. 
The widening of the opening at the 
back, converts ee, ai, ae, into i, e, a. 
Now e is much finer than ae, and re- 
places it in the South of England. 
Care must be taken not to confuse 
English a with aa. The true a seems 
almost peculiar to the Southern and 
Western, the refined Northern, and 
the Irish pronunciation of English. 
The exact boundaries of the illiterate 
a and aa have to be ascertained. 
Eounding the lips changes ee, ai, ae, 
into ni, eo, ed , of which eo is very 
common. Rounding the lips also 
changes i, e, a, into ue, oe, oe\ of which 
oe is very common. 

On uttering oo, oa, au, the back of 
the tongue descends lower and lower, 
till for ail the tongue lies almost en- 
tirely in the lower jaw. The widening 
of these gives uo, ao, o. The distinction . 
between an, o, is necessarily very slight ; 
as is also that between ao and o. But 
ao is very common in our dialects, and 
is known as o aperto in Italy. The 
primary forms of oo, oa, au, produced 
by opening the lips, are the obscure 
uu' , uu, ua, of which uu is very common 
in the provinces, being a deeper, thicker, 
broader sound of u. But the wide 
sounds uo, ao, o, on opening the lips, 
produce u\ aa, ah. Here aa is the 
true Italian and Spanish a, and ah is 
the deeper sound, heard for long a in 
Scotland and Germany, often confused 
with th3 rounded form au. 

Of the mixed vowels, the only im- 
portant primary vowel is u, for which 
the tongue lies flat, half way between 
the upper and lower jaw. It is as 
colourless as possible. It usually re- 
places uu in unaccented syllables, and 
altogether replaces it in refined South- 
ern speech. Its wide form a' is the 
modern French fine a, much used also 
for aa in the South of England. The 
rounded form oa' seems to replace u or 
nu in some dialects. The mixed sound 
resulting from attempting to utter ah 
and a together is e', which Mr. Bell 
considers to be the true vowel in herd. 

Distinctions to be carefully drawn in 

writing dialects. EE and I. AI and 
E. AE and E. AA, AH and A. 
OA and AO. AO, AU and AH. 00 
and UO. UU and U. UI, UE and 
FEW, IW, YOO. U£ and EO. 
OE and U. 

Quantity of Vowels. 

All vowels are to be read short, or 
medial, except otherwise marked. 

The Stress (•) placed immediately after 
a vowel shews it to be long and ac- 
cented, as au'gust ; placed immedi- 
ately after a consonant, hj^ihen (-), 
gap' (:), or stop (..), it shews that 
the preceding vowel is short and ac- 
cented, as augus't, aamao.', pa'pa'..' 

The Holder (••) placed immediately 
after a vowel or consonant shews it 
to be long, a^ aicgus't, needl" ; the 
Stress Holder (•••) shews that the 
consonant it follows, is held, the pre- 
ceding vowel being short and accent- 
ed, compare hap-i, hap"'i, ha'pi, 
ha'p"i ; in theoretical writing only. 
Practically it is more convenient to 
double a held consonant, as hap'i, 
kap'pi, ha'pjn. 

Stop (..) subjoined to any letter indi- 
cates a caught-up, imperfect utter- 
ance, as ka.., kat.. for kat ; great 
abruptness is marked by (...) 

Accent marks may also be used when 
preferred, being placed over the first 
letter of a combination, thus : 

with stress — ua" 
without stress — aa" 







aa aa aa 

If the first letter is a capital the accent 
marks may be placed on the second, 
as August, augiist, kdazda. 

Systematic Diphthongs. 

The stressless element of a diph- 
thong is systematically indicated by a 
preceding turned comma (') called 
hook, as m^eeai^ee It. miei, Laa^ooraa 
It. Laura, p^aaoo'raa It paura, I'-ueee 
Fr. lui. But when, as is almost always 
the case, this element is '•ee 'oo, or 'ue, 
it may be replaced by its related con- 
sonant g, w or ^10, as mgaiy, Laawraa, 
l^wee. Any obscure final element as 
% 'e, 'e\ is sufficiently expressed by 
the sign of simple voice h\ as provin- 
cial neeKt night, streeKm stream 
wiKkn waken. In applying the rule 
for marking stress and quantity, treat 
the stressless element as a consonant. 



The four English Glossic diphthongs 
EI, 01, ou, EU are unsystematic, and 
are variously pronounced, thus : 
EI is uy in the South, sometimes a'y, 
aay ; and is often broadened to imy^ 
ahy, atiy, in the provinces. 
01 is oy in the South, and becomes any, 

ou is uw in the South, sometimes a'to, 
aaw, and is often broadened to uuw 
ahw^ oaw, aow ; it becomes oe^w in 
Devonshire, and aew in Norfolk. 
EU varies as iw^ eew^ yoo, yiiv, yeew. 

The Londoners often mispronounce 
Ai as ai'y, aiy, ey or nearly uy, and oa 
as oa'w, oaw, otv or nearly mv. 

English vocal r, is essentially the 
same as H', forming a diphthong with 
the preceding vowel. Thus English 
glossic peer, pair, boar, boor, fer, difer'- 
ring, are systematic ^r A', pe'h\ bao'h\ 
buo-}i\ fe'h' or fw, dife'h'-ring or 
dif wring. But r is used where r', or 
rr\ or AV may be occasionally heard. 


Differences from English Glossic con- 
sonants are marked by adding an h in 
the usual way, with y for pnlatals, 
and w' for labials, by subjoining an 
apostrophe ( ' ) or by prefixing a turned 
comma ('), a turned apostrophe ( ^ ), 
or a simple comma (,). 

Simple consonants, and added G. 

Y, W, H ; P B, T D, J, K G, F V, S Z, 
vocal E,, L M N, NG. 

Added H. 


KH, GH German ch, g in Bach, Tage ; 

are the hissed voiceless forms of 
y, r', I, m, n, ng. 

Added Y' and TIT. 

TY', DY', KY',GY', LY', NY', NGY', 

are palatalised or mouille yavieties 
of i, d, k, g, I, n, ng, as in virtue, 
verdure, old cart, old guard, Italian 
gl, gn, vulgar French, il ny a 
pas=ngy'aa pah. LYH is the 
hissed voiceless form of LY'. 
KYH, GYII are palatal varieties of 
KH, GH as in German ich, fliege. 

Added W and WB. 

TW, DW, KW, GW, EW, R'W, 
LW, NW, &c., are labial varieties 

of t, d, Jc, g, r, r\ I, n, &c., pro- 
duced by rounding the lips at or 
during their utterance, French toi, 
dois, English quiet, guano, our, 
French roi, hi, noix, &c. 
KWH, GWH are labial varieties of 
KH, GH as in German auch, saugen, 
and Scotch quh. HWH is a whistle. 

Added apostrophe (') called " SooJc.'" 

H' called aich-huok,\?, the simplestemis- 
sion of voice: H'W is A' with round- 
ed lips ; H' WH a voiced whistle. 

T', D', called tee-huok, dce-huok, dental 
t, d, with tip of tongue nearly 
between teeth as for th, dh. 

F', V, called cf-huok, vee-huok, tooth- 
less /, V, the lip not touching the 
teeth ; v' is true German iv. 

r', or R before vowels, is trilled r. 

N' read en-Jnioh, French nasal n, which 
nasalizes the preceding vowel. 'J'o 
Englishmen the four French words 
vent, vont, vin, un sound von\ voan\ 
van', un' ; but Frenchmen take 
them as vahn', voan' , vaen' , oen' . 
Sanscrit unuosvaa^ru. 

K', G' peculiar Picard varieties of 
kif, gy'. nearly approaching ch, j. 

CH', J', TS', DZ' monophthongal 
Eoman varieties of ch, j, ts, dz. 

T'H, D'H lisped varieties of 5, z, imi- 
tating th, dh ; occasional Spanish 
z, d. 

S' not after t, Sanscrit visu^rgu. 

Frefixed comma (,), called " Comma." 

,H read koma-aich, lax utterance, op- 
posed to .H. 

,T ,D read koma-tee, koma-dee peculiar 
Sardinian varieties of t, d, the 
tongue being much retracted. 

,L Polish barred I, with ,LH its voice- 
less, ,LW its labial, and ,LWH 
its voiceless labial forms. 
; read hamza, check of the glottis. 

Prejlxed turned comma ('), called 

' read ein, the Arabic '.aayn or bleat. 

'H, 'T 'D, 'S 'Z, 'K, read huok-aich, 
huok-tee, &c. ; peculiar Arabic 
varieties of h, t, d, s, z, k; 'G the 
voiced form of 'K. 

'KH, *GH, called huok-kai-aich, huok- 
jee-aich ; the Arabic kh, gh pro- 
nounced with a rattle of the uvula. 



'W, 'PR, 'BR, read hnoh-diibl-eu, &c.; 
lip trills, the first with tif^ht and 
the others with loose lips ; the first 
is the common English defective w 
for r', as vehvi Vivoo , the last is 
used for stopping horses in Germany. 

*R read huok-am\ the French r grasseye, 
and Northumberland burr or k^ruop 
= ^gh\. ; 'EH its voiceless form. 

'LII, 'L, read huok-el-aich^ huok-el, 
"Welsh //, and its voiced Manx form. 

'F, 'V, read hiiok-ef &c. ; /, v with back 
of tongue raised as for oo. 

Prejixed turned apostrophe (J, called 
" Curve:' 

,AA, read kerv-aa, an aa pronounced 
through the nose, as in many parts 
of Germany and America, different 
from flflw', and so for any vowel, 
°h, or h'. 

.T ,D, .SH, .R, .L, ,N read kerv-tee &c., 
Sanscrit "cereWal" t, d, sh, r\ l,n; 
produced by turning the under part 
of the tongue to the roof of the 
mouth and attempting to utter t, d, 
sh, r\ /, n. 

,H read kerv-aich, a post aspiration, 
consisting of the emphatic utter- 
ance of the following vowel, in one 
syllable with the consonant, or an 
emphatically added final aspirate 
after a consonant. Common in 
Irish-English, and Hindoostaanee. 

^"W is the consonant related to tie, as 
w is to 00. 

Clicks, — spoken with suction stopped. 

C, tongue in t position, English tut ! 

Q, tongue in / position. 

X, tongue in ty position, but unilateral, 
that is, with the left edge clinging 
to the palate, and the right free, as 
in English clicking to a horse. C, 
q, X, are used in Appleyard's Caff re. 

0,0, tongue in ty position, but not 
unilateral ; from Boyce's Hottentot. 

KG, tongue retracted to the ^k position 
and clinging to the soft palate. 

Whispers or Flats. 

°H, called serkl-aich, simple whisper; 
°H' whisper and voice together 
«°H' diphthongal form of ^^A'. 

°AA, read serkl-aa, whispered aa, and 
so for all vowels. 

°B, °D, read serkl-bee etc., the sound of 
b, d, heard when whispering, as dis- 
tinct from p, t, common in Saxony 
when initial, and sounding to 

Englishmen like p, t when stand- 
ing for b, d, and like b, d when 
standing for p, t. °G, whispered g^ 
does not occur in Saxony. 
°V, °DH, °Z, °ZH, °L, °M, °N read 
serkl-vee etc., similar theoretical 
English varieties, final, or interposed 
between voiced and voiceless letters. 


The tones should be placed after the 
Chinese word or the English syllable 
to which they refer. They are here, 
for convenience, printed over or un- 
der the vowel o, but in writing and 
printing the vowel should be cut out. 
0, 0, high or low level tone, pjiing~. 
o, o, tone rising from high or low pitch, 

6, o rise and fall, (that is, foo-kyen 

shaang\) or fall and rise. 
6, o falling tone to high or low pitch, 

kyoo^ or kjioe'. 
o^ n sudden catch of the voice at a 
high or low pitch, shoo^, zhee^'^ 
nyip" , or yaap . 


Hyphen (-), used to separate combina- 
tions, as in mis-hap, in-got. In 
ivhair-ever, r is vocal ; elm fauln 
are monosyllables, el-m, faul-n are - 
dissyllables ; Jidler has two syllables, ■ 
fidl-er three syllables. ■ 

Divider ), occasionally used to assist 
the reader by separating to the eye, 
words not separated to the ear, as 
teT)er dhat)l doo. 

Omission (J, occasionally used to assist 
the reader by indicating the omission 
of some letters usually pronounced, 
as hee)J, do6)jt. 

Gap (:) indicates an hiatus. 

Closure (.) prefixed to any letter indi- 
cates a very emphatic utterance as 
mei .hei for my eye. 

Emphasis (•) prefixed to a word, shews 
that the whole word is more em- 
phatically uttered, as ei 'neu dhat 
'dhat dhat 'dhat man sed woz rong ; 
'ei gaiv 'too thingz too 'too men, and 
'hee gaiv 'too, 'too, too 'too, 'too. 

The following are subjoined to indicate, 
\ emission, j suction, ^ trill of the 
organs implicated, f inner and 4- 
outer position of the organs impli- 
cated, % tongue protruded, § unilate- 
rality, * linking of the two letters 
between which it stands to form a 
third sound, ( extreme faintness. 



*#* The Eeader should pay particular attention to the Rules for marking vowel 

quantity laid down in the Key, p. xvi. 

EoEEiGi^ Languages. 
French. — Ai p^wee uen vyaiy ka'raony' ai iin'n)on'foii' bao'my' 
oan' von'due deo moavae van' oa poeplh bae*"t. Ee aet voo ? 

German. — Ahkh ! aaynu' aayntseegyhu' ue'blu' foyreegyhu' 
mueku' koentu' v'oal ahwkwh meekyh boe'zu' mahkhu'n ! Yhah* 
szoa* ! Es too't meer' oonien'dleekyli laayt ! 

Old English. 
Conjectured Pronunciation of Chaucer, transliterated from ^^ Early 
English Pronunciation,^'' p. 681 : 

"Whaan dhaat Aa-pri"l with)is shoo -res swao'te 
Dhe droo'kwht aof Maarch haath per'sed tao dhe rao'te, 
Aand baa'dhed evri* vaayn in swich li'kooT 
Aof which ver'tue* enjen'dred is dhe floo'r; 
Whaan Zefiroos, e"k, with)is swe'te bre'the 
Inspi-red haath in evri* haolt aand he*the 
Dhe tendre kropes, aand dhe yoonge soone 
Haath in dhe Eaam is)haalfe koo-r's iroon'e, 
Aand smaa'le foo'les maa'ken melaodi'e, 
Dhaat sle-pen aal dhe nikyht with ao-pen i'e, — 
Sao priketh hem naa'tueT in her' kao'raa'jes; 
Dhaan laongen faolk tao gao'n aon pil'gri-maa'jes, 
Aand paalmerz faor' tao se*ken straawnje straondes, 
Tao fer'ne haalwes koo'th in soon'dri* laondes ; 
Aand spes'iaali* fraom evri* shi'res ende 
Aof Engelaond, tao Kaawn*ter'ber*i* dhaay wende, 
Dhe hao'li* blisfool maar'"tiT faor tao se*ke, 
Dhaat hem haath haolpen, whaan dhaat dhaay wot se*ke. 
Dialectic English and Scotch. 
Received Pronunciation. — Whot d)yoo wont? Vulgar Cockney. — 
Wau'chi wau'nt? Devonshire. — Wat d)yae want? Fifeshire. — 
"Whuu't u'r' yi' waan;n ? Teviotdale. — Kwhaht er' ee wahntun ? 
Teviotdale, from the dictation of Mr. Hurray of Hawick. — Dhe)r' 
ti'wkwh sahkwhs graowun e dhe Ri'wkwh Hi'wkwh Hahkwh. 
— Kwhaht er' ee ah'nd um ? U')m ah'ndum naokwht. — Yuuw un 
•mey el gu'ng aowr' dhe deyk nn puuw e pey e dhe muunth e 
Mai'y. — Hey)l bey aowr' dhe 'naow nuuw. 

Aberdeen. — Faat foa*r' di'd dhe peer' si'n vreet tl)z mi'dher' ? 
Glasgow. — Wu)l ait wur' bred n buu;ur' doon dhu waa;ur'. 
Lothian. — Mahh' koanshuns ! hahng u' Be*yli ! — Gaang u'wah*, 
laadi ! gai tu dhu hoar's, sai xx ! un shoo em 'baak ugi'n* ! 

Norfolk. — Wuuy dao-nt yu' paa')mi dhaat dhur -tue paewnd yu' 
ao")mi, bo ? TJuy dao'nt ao*)yu' nao 'tue paewnd. Yuuw 'due ! 

Scoring Sheep in the Yorkshire Dales. — 1. yaan, 2 taih'n, 3 tedh- 
uru, 4 medhuru (edhuru), 5 pimp (pip), 6 saa'jis (see'zu), 7 laa'jis 
(re-ru), 8 sao'va (koturu), 9 dao'vu (hau'nu), 10 dik, 11 yaan 
uboo"n, 12 tain uboo*n, 13 tedhur' uboo'n, 14 medhur' uboon, 
15 jigit, 16 yaan ugeeh'-n, 17 tain ugeeh'-n, 18 tedhur' ugeeh'-n, 
19 medhur' ugeeh'*n, 20 gin ageeh'-n (bumfit). 



Dialects of the Peak of DERi^YsniRE from the dictation of 
Mr. Tuomas Halla^i, of Manchester, a native of the Peak. 

*^* Mr. TIallam considers that he said a\ uo. uow, vdeys, where I seemed to hear 
and vvrotu aa, oa", uV iv, va ys. Mr. ilallani dictated the quantities. 


Th)Sdangg u) Solumun, Chdapt'ur th)- Th)Sda'ngg u)S6luimm^ Chdaptur ih)- 
sdckund. sdekund. 

1. Ai5)m th)roaz u)Shaeruii un)th)- 
lilli u)th vaalliz. 

2. Lahyk th)lini umoa'ng thaurnz, 
sui'w iz maby liiuv uraoa'ng th)- 

3. Lahyk th)imppl t'riy umoa'ng 
th)t'riyz u)th wua'd, sui'w iz mahy 
biluuvd uraoa'ng th)s6a'nz. Au sit)mi 
daawn wi gract dliy 6a'nd'ur')iz 
shaadu, un)iz)lrui'wt wur)swiyt tu)mi 

4. ly bruuwt)mi tu)th)feeh'stin 
aaws, un)iz)fla'g oar mi wur luuv. 

5. St'raengthn)mi "wi)soa'mut** 
d'ringk, kuumfurt)rai wi)aapplz : fur 
au)m luuv-sik. 

6. Iz lift 6nt)s oa'nd'ur mi)yaed, 
im)iz riyt 6nt tlips)mi. , 

7. Au chaarj)yu, oadiiuwt't'rz u)Ji- 
Ttii'wslum, bi)th)r6az, un)bi)tn)sta'gz 
u)th)fiylt, uz yoa mun noadhur stuur, 
nur wa'kn mi)luuv, til)iy)pleeh'zuz. 

8. Th)va'ys u)mi)biluuvd ! Lui'wk, 
iykuumz leeh'pin oa'pu)th)maawntinz, 
sky'ippin 6a'pu)th ilz. 

9. Mi)biluuvd)z lahyk u)roa, ur')u)- 
yoa'ng sta'g : liii'wk, iy stondz ut)- 
ba'k)u aar)wau, iy lui'wks aawt ut)- 
th)windus, uu)shoaz issael thrui'w)- 

10. Mi) biluuvd spuuk, un)saed 
tCii'w)mi, Gy'aet oa'p, mi)lLiuv, mi)- 
faer')un, un)kuum uwai. 

11. Fur, lui'wk, th)wint'ur)z paast, 
iin)th)rain)z oar un)gaun. 

12. Th)flaawurz ur)kuurain oa'pu)- 
th) graawnd, th)taliym)z kuuran us)th)- 
bridz singn, un)th)va'ys u)th)tuurtl)z 
eerd i)aarjk6a'nt;'ri. 

13. Th)fig t'riyz ur) gy'aetin griyn 
figz on, un)tli)vabynz gy'in u)nc\hys 
smael wi)tli)y6a'ng graips. Gy'aet 
oa'p, mi)luuv, mi)faer*)un, un)kuum 

14. Oa mahy doav, uz)urt)i)th)tlifs 
u)th)v6k, i)th)saikrit spots u)th) staerz, 
lae)mi siy dhijfais, lae)mi eer dhi)- 
va'ys; fur)dhi) va'ys is swiyt, un)dhi)- 
fais iz vaerri praati. 

1. A<'i)m th)roaz u)Shaerun un)th)- 
lilli u)th vaalliz. 

2. Us th)lilli umoa'ng thiumz, soo 
iz mau liiuv umoa'ng th)duuwtturz. 

3. Us th)aappl traey umoa'ng th)- 
traeyz u)th woa'd, s6o)z mau biluuvd 
umoa'ng th)s6a'nz. Aii sit daawn wi 
greet dlaey da'ndur')iz shaadu, un)iz)- 
fri'wt wur)swaeyt tu) mi) taist. 

4. Aey brtiuwt)mi tu)th)feestin aaws, 
un)iz)fla'g 6ar)mi wur liiuv. 

0. Ky'aeyp mi oa'p wi' soa'mut" 
dringk, kuumfurt)mi wi)aapplz ; fur 
au)m luav-sik. 

6 Iz lift 6nd)z oa'ndur mi)yaed, un)- 
iz raeyt ond tlips)mi. 

7. Au tae])vu, 6a duuwtturz u)Ji- 
rtiuwslum, bi)th roaz, un)bi)th)sta'gz 
u)th faeylfc, dhut yda mun noadhur stiiur 
nur waakn mau luuv, til aey lahyks. 

8. Th)vai<ys ujmi) biluuvd! Luuwk, 
aey kuumz le'eppiu 6a pu)th)maawn- 
tinz, sky'ippin 6a'pu)th ilz. 

9. Mi)biluuvd)z lahyk u)r6a, ur')u)- 
yoa'ng sta'g : luuwk, aey stondz ut)- 
th)baak)u aar)wau, aey liiuwks aawt 
ut)th)windus, un)!5hoaz issael thriiuw)- 

10. Mi)biluuvd spauk, un)s^ed 
ttiuw)mi, Gy'aer')6a'p, mi)luuv, mi). 
faer')un, un)kuum uwee. 

11. Fur, liiuwk, th)wintur)z paast, 
un)th)reon)z oar un)gaun. 

12. Thjflaawurz ur)kiiumin oa'pu)- 
th) graawnd, th)tah}Tn)z kiiumn us)th)- 
bridz singn, un)th)vahys u)th)tuurtl)z 
eerd i)aar)k6a'ntri. 

1 3. Th)f ig traeyz ur)gy'aetin graeya 
figz on, un)th)vahynz gy'in ujnahys 
smf\el wi)th)y6a'ng graips. Gy'aer')- 
oa'p, mi)luuv, mi)faer')un, un)kiium 

14. Oa mau doav, uz)urt)i)th)niks 
u)th)rok, i)th)seekrit spots u)th)staerz, 
lae)mi saey dhi)fais, lae)mi eer dhi)- 
vahys; fur)dhi)vahys is swaeyt, un)- 
dhi)fais iz vaerri praati. 

Separate Copies of this Notice and Appendix on Glossic will he 
sent on application to the Author. 



Illustrations of the Pronunciation of English during 
THE Fourteenth Century. 

§ 1. Chaucer. 

Critical Text of Prologue. 

In accordance with, the intimation on p. 398, the Prologue 
to the Canterbury Tales is here given as an illustration of 
the conclusions arrived at in Chap. lY., for the pronuncia- 
tion of English in the xiv th century. But it has been 
necessary to abandon the intention there expressed, of follow- 
ing the Harl. MS. 7334 as closely as possible, for since the 
passage referred to was printed, the Chaucer Society has 
issued its magnificent Six- Text Edition of the Prologue and 
Knight's Tale, and it was therefore necessary to study those 
MSS. with a view to arriving at a satisfactory text to pro- 
nounce, that is, one which satisfied the laws of grammar and 
the laws of metre better than the reading of any one single 
MS. which we possess. For this purpose the systematic 
orthography proposed on p. 401, became of importance. The 
value of exact diplomatic reprints of the MSS. on which we 
rely, cannot be overrated. But when we possess these, and 
endeavour to divine an original text whence they may have 
all arisen, we ought not to attempt to do so by the patch- 
work process of fitting together words taken from different 
MSS., each retaining the peculiar and often provincial or- 
thography of the originals. The result of such a process 
could not but be more unlike what Chaucer wrote than any 
systematic orthography. Chaucer no doubt did not spell 
uniformly. It is very difficult to do so, as I can attest, after 
making the following attempt, and probably not succeeding. 
But a modern should not venture to vary his orthography 
according to his own feelings at the moment, as they would 
be almost sure to lead him astray. Whenever, therefore, a 
text is made out of other texts some sort of systematic ortho- 
graphy is inevitable, and hence, notwithstanding the vehe- 


634 LONG U IN SEVEN MSS. Chap. VII. § 1. 

ment denunciation of the editor of the Six-Text Edition/ 
I have made trial of that one proposed on p. 401, in all its 
strictness. The result is on the whole, better than could 
have been expected. Notwithstanding the substantial agree- 
ment of the Harleian 7334, and the Six New Texts, there is 
just sufficient discrepancy to assist in removing almost every 
difficulty of language and metre, so far as the prologue is 
concerned, and to render conjecture almost unnecessary. 
The details are briefly given in the footnotes to the following 
composite text. 

Pronunciation of Long U and of AY, EY as deduced from a comparison 
OF THE Orthographies of Seven Manuscripts of the Canterbury 

The investigations in Chap. lY. for the determination of the pro- 
nunciation of the XIV th century, were avowedly founded upon the 
single MS. Harl. 7334 (supra p. 244). 'Novf that large portions 
of six other MSS. have been diplomatically printed, it is satisfactory 
to see that this determination is practically unaffected by the new 
orthogi'aphies introduced. The Cambridge and the Lansdowne 
MSS., indeed, present us at first sight with what appears to be 
great vagaries, but when we have once recognized these as being, 
not indeterminate spellings of southern sounds, but sufficiently 
determinate representations of provincial, northern, or west midland, 
utterances, mixed with some attempts to give southern pronuncia- 
tion, they at once corroborate, instead of invalidating, the conclu- 
sions already obtained. That this is the proper view has been 
sufficiently shewn in the Temporary Preface to the Six-Text 
Edition, p. 51 and p. 62, and there is no need to discuss it further. 

1 Temporary Preface to the Six- the editor's track, and often stand in 
Text Edition of Chaucer's Canterbury the way of an independent conjecture. 
Tales, Part I., by F. J. Furnivall, pp. At the same time they do not present 
113-115. A uniform system of spell- the text as the editor would shew it, 
ing did not prevail in the xiv th cen- for the attention is distracted by the 
tury, and as we have seen, can scarcely brackets. The plan pursued for the 
be said to prevail in the xixth, but Prisoner's Prayer, supra pp. 434-437, 
variations were not intentional, and the of giving the original and amended 
plan I advocate is, from the varied texts in parallel columns, is the only 
spellings which prevail, to discover the one which fully answers both pur- 
system aimed at, but missed, by the old poses. Where this is not possible, it 
writer, and adopt it. All varieties of it appears to me that the best course 
grammar, dialect, and pronunciation, to pursue is to leave the text pure, and 
when belonging to the author, and not submit the correction in a note. This 
his scribe, who was often ignorant, and serves the purpose of the [ ] or sic^ 
still oftener careless (p. 249), should be much more eflfectually than such dis- 
preserved, and autographs, such as turbances of the text, which are only 
Orrmin's and Dan Michel's, must be indispensable when notes are incon- 
followed implicitly and literatim. In venient. The division of words and 
such diplomatic printing, I even object capitals of the original should for the 
to insertions between brackets. They same reason be retained. See the 
destroy the appearance of the original. Temp. Pref. p. 88. 
and hence throw the investigator into 

Chap. YII. § 1. LONG U IN SEVEN MSS. 635 

These MSS. may be looked upon as authorities for the words, but 
not for the southern pronunciation of the words, and they shew their 
writers' own pronunciation by using letters in precisely the same 
sense as was assigned from the Harl. MS. on p. 398 above. Two 
points may be particularly noticed because they are both points of 
difference between Mr. Payne and myself, (supra pp. 582, 583) 
and in one of them I seem to differ from many of those who have 
formed an opinion on the subject. 

Long u after an examination of all the authorities I could find, 
was stated on p. 171 to have been (yy) during the xvith century. 
There did not appear to be any ground for supposing it to be 
different in the xivth century, and hence it was assumed on 
p. 298 to have had that value at that time. This was strengthened 
by the proof that (uu), the only other sound which it could 
have represented, was written ou, p. 305. A further though a 
negative proof seems to be furnished by the fact that I have 
not observed any case of long u and ou rhyming together, or 
being substituted one for the other in the old or any one of the 
six newly published texts. ^ I cannot pretend to have carefully 
examined them for that purpose, but it is not likely that in my 
frequent references to them for other purposes, such a marked 
peculiarity should have escaped me. It has however been already 
pointed out that in the first half of the xnith century (uu) was 
represented by u, and not by ow, and for about thirty years, includ- 
ing the end of the xiii th and beginning of the xiv th century, both 
signs were employed indiscriminately for (uu), and that this use of 
ou seemed to have arisen from a growing use of u as (yy), pp. 424, 
470, 471 note 2, etc.^ Hence the predominance of ou in the be- 

^ Compare fortone, hike in Hampole Judging however by the collation in 
(supra p. 410, n. 2). The two ortho- F. Michel's edn. the Oxf. MS. agrees 
graphics boke, buke, struggle with each with the Cam. The text is clearly- 
other in Hampole. In the Toivneley doubtful. 

Mysteries, I have also observed the But v. 691, which in the Cam. MS. 

rhyme, goode infude, which however, runs 

may be simply a bad rhyme, the spell- he lij? in bure 

ing is Northern and of the latter part under cou<?/-ture 

of the XV th century. On examining becomes in the Harl. fo. 87, 

the Harl. MS. 2253 for the rhymes : he byht nou in boure, 

bur mesaventur, bure coverture, quoted vnder couertoure, 

from the Cam. MS. of King Horn on where the scribe by adopting the or- 

p. 480, I find that the first rhyme dis- thography ou has clearly committed 

appears. Thus v. 325, Lumby's edition himself to the pronunciation (uu) and 

of the Cam. MSS. has not (yy). It would, however, not be 

Went ut of my bur safe to draw a general conclusion from 

Wi]7 muchel mefaventur these examples in evidently very un- 

and the Harl. reads fo. 85, trustworthy texts, which have yet to 

Went out of my boure, be properly studied in connection with 

fhame ]>& mott byfhoure ; dialectic and individual pronunciation, 

and V. 649, the Cam. MS. has supra p. 481. 

heo ferde in to bure ^ On p. 301, note, col. 1, a few in- 
to fen aue>^t?«-e, stances of the Devonshire substitutes 
and the Harl. has, fo. 87, for (uu) are given, on the authority of 
Horn ne ]7ohte nout him on Mr. Shelly's pronunciation of Nathan 
ant to boure wes ygon. Hogg's Letters. The new series of 



Chap. VII. § 1. 

ginning of the xiv th century and the subsequent strict severance of 
long u and ou, which seem so far as I have observed, to have been 
never confused, as short u and ou certainly were (p. 304). The 
conclusion seems to be inevitable, that long u and ou represented 
different sounds, and that the long u must have had in the xiv th, 
whatBuUokar in the xvith century called its '' olde and continued'* 
sound, namely (yy). This, however, is directly opposed to Mr. 
Payne's opinions given on p. 583. 

those letters there named, having an 
improved orthography, using u, a, for 
(y, 8b), — not (a), as there misprinted, — 
has allowed me to make some collec- 
tions of words, which are curious in 
connection with the very ancient west- 
ern confusion of u, e, i, and the pro- 
nunciation of long u as (yy). It may 
be stated that the sound is not always 
exactly (yy). In various mouths, and 
even in the same mouth, it varies 
considerably, inclining towards (uu), 
through (uu?), or towards {99) the labi- 
alised {ee). The short sound in did 
seemed truly (d^d). But in could, goody 
I heard very distinctly (kyd, gyd) with 
a clear, but extremely short (y), from 
South Devon peasants in the neigh- 
bourhood of Totnes. Nor is the use of 
(yy) or (uu, 99) for (uu) due to any in- 
capacity on the part of the speaker to 
say (uu). The same peasant who 
called Combs, (KyjTuz) or (K^smz), 
[it is difficult to say which, and appa- 
rently the sound was not determinate], 
and even echoed the name thus when 
put to him as (Kuumz), and called brook 
(bryk), with a very short (y), talked 
of (muur, stuunz, ruud) for 7nore, stones, 
road. Mr. Murray, in his paper on 
the Scotch dialect in the Philological 
Transactions, has some interesting spe- 
culations on similar confusions in 
Scotch, and on the transition of (u) or 
{u) through {9) into (a) and finally (a). 
On referring to pp. 160-3, supra, the 
close connection of (uu, yy) will be seen 
to be due to the fact that both are 
labial, and that in both the tongue is 
raised, the back for (uu) and front 
for (yy). The passage from (uu) 
to (yy) may therefore be made almost 
imperceptibly, and if the front is 
slightly lowered, the result becomes 
{99). The two sounds (yy, 99) are 
consequently greatly confused by 
speakers in Scotland, Norfolk, and 
Devonshire. Mr. Murray notes the 
resemblance between {9, o), — which in- 
deed led to the similarity of their nota- 

tion in palaeotype — as shewn by Mr. 
M. Bell's assigning (a) and my giving 
{9) to the French mute e, which others 
again make (3»h). If then (u) travels 
through (y, 9) to (a), its change to (a) 
is almost imperceptible, and the slight- 
est labialisation of the latter sound 
gives (0). Whatever be the reason, 
there can be no doubt of the fact that 
(u, y, 9, a, a, 0) do interchange pro- 
vincially now, and hence we must not 
be surprised at finding that they did 
so in ancient times, when the circum- 
stances were only more favourable to 
varieties of speech. These observations 
will serve in some degree to explain 
the phenomena alluded to in the text, 
and also the following lists fi-om Nathan 
Hogg's second series, in which I re- 
tain the orthography of the author 
(Mr. H. Baird), where we should read 
u, a as (y, ae) short or long, and other 
letters nearly as in glossotj'pe. 

EW and long U become (yy) , as : 
h\u, bwty, cruel, cwryiss curious, cuty 
acute, duce deuce, de^ty, hw hue yeWy 
hwmin humany kinklwd conclude, muzia, 
nu new, pwr pure, redn'd, stw steWy 
stwpid, iru, tr«<th, tun, vlwt flute, \u 
view few, vwm fume, vwtur futurCy 
y««z'd used, zwant suant. 

Long and short 00, OIJ, 0, U, 
usually called (uu, u) become (yy, y) or 
{99, 5>), as: balw hullahbaloo,\Aum. bloom, 
brwk brook, bwk book, ch»<z choose, crwk 
crook, cwd could, cuxi court, ce<s course 
coarse, dmu through, dr//pin droopingy 
dw do, gwd good, g«Jden golden, intw, 
k^^shin cushion, luklook, k^s'nd loosened, 
min^^ver manoeuvre, mwv move, nun. 
noon, ^uVdi pulled, prwv prove, pwk 
pook, xum. room, sku shoe, shwd should, 
skade school, stwd stood, tropin trooping, 
ill too two to [emphatic, unemphatic 
ta = (t9)], tu\ took, twm tomb, u whoy 
Yu\ full fool, ve^t foot, ju you, zmwthe 
smooth, zun soon. 

Short U, 00, usually called (a) 
become (i), as : blid blood, dist do'st, 
honjist, unjusty jist just adv., rin run 

Chap. VII. § 1. AI AY, EI EY, TN SEVEN MSS. 637 

The second point is extremely difficult, and cannot be so cursorily 
dismissed. What was the sound attributed to ai «y, et ey in 
Chaucer ? The constant confusion of all four spellings shews that 
it was one and the same.^ Here again the voice of the xvith 
century was all but unanimous for (ai), but there is one remarkable 
exception, Hart, who as early as 1551 (in his MS. cited below 
Chap. VIII, § 3, note 1), distinctly asserts the identity of the 
sounds of these combinations with that of e, ea, that is (ee). For 
printing this assertion in 1569 he was strictly called to order by 
Gill in 1621, supra p. 122. All the other writers of thexvith 
century, especially Salesbury and Smith distinctly assert that (ai) 
was the sound. Hence on p. 263, (ai) was taken without hesitation 
to be the sound of ay, ey, in Chaucer. We are familiar with the 
change of (ai) into (ee), p. 238, and with the change of (ii) into (ai, 
ai), p. 295, but the change of (ee) into (ai), although possible, and 
in actual living English progress (p. 454, n. 1), is not usual. 
There was no reason at all to suppose that ay could have been (ii), 
and little reason to suppose that it would have been (ee) before it 
became (ai). On examining the origin of ay, ey, in English words 
derived from ags. sources, the y or i appears as the relic of a former 
g = (gh, g\ j) and then (i), which leads irrresistibly to the notion 
of the diphthong (ai), p. 440, 1. 14, p. 489. But it certainly does 
not always so arise, and we have seen in Orrmin (ib.) that the 
jj = (j) was sometimes as pure an insertion as we occasionally 
find in romance words derived from the Latin, ^ and as we now find 

[also to urn'], rish'd rushed, tich'd ^ Not in Scotch, where the spellings 

touched, vlid flood, wid'n would not, ai, ei seem to have been developed in- 

winder wonder, wisser worser, zich dependently in the xv th century, for 

such^ zin sun son, zmitch smutch. the Scotch long a, e, and perhaps 

Short E, I, usually called (e, i) are meant (a-B, ee), compare Sir T. Smith, 

frequently replaced hy (a) or (a), as : supra p. 121, 1. 18. These spellings 

bevul befell, hul bell, bulch'd belched, were accompanied by the similar forms 

burry'd buried, churish cherish, eszul oi, ui, oui for the long o, u, ou, per- 

himself, etszul itself, mezul myself, hap8 = (oB, ye, u-e), though the first 

mulkin milking, muller miller, purish was not much used. "We must recol- 

perish, shullins shillings, spul spell, lect that in Scotch short i was not (i) 

spurrit spzVzY [common even in London, or {i), but (e), and hence might easily 

and compare syrop, stirrup'], tullee tell be used for (b) or (a) into which un- 

you, turrabul terrible, ulbaw'd elboived, accented (e) readily degenerates. For 

vuller fellow [no r pronounced, final or this information I am indebted to Mr. 

pre-consonantal trilled (r) seems un- Murray's paper on Scotch (referred to 

known in Devonshire], vullidge v^7^a^(^, in the last note), which was kindly 

vulty filthy, vurrit ferret, vury very, shewn to me in the MS. The notes 

yvisi first, wul well, wulvare welfare, yul there furnished on the development of 

yell, }Tir'd heard, zmul smell, zulf self Scotch orthography are highly interest- 

The words ze^'d swept, indwd indeed, ing, and tend to establish an intentional 

dwd did done, humman hummen woman phonetic reformation at this early 

women, do not exactly belong to any period, removing Scotch spelling from 

of these categories, the historical affiliation which marks 

The above lists, which, being only the English, 

derived from one small book, are ne- ^ " In Italian, Spanish, Portuguese, 

cessarily very incomplete, serve to shew and Provencal, Latin A remains un- 

the importance of modern dialectic altered. Some deviations into ai or e 

study in the appreciation of ancient must be admitted. . . . The most im- 

and therefore dialectic English (p. 581). portant and frequent case is when a by 



Chap. VII. § 1. 

iu EnglLsh after the sound of {ee) in what many persons recognize 
as the "standard" pronunciation of our language, for instance 
(ne^im) for name. There are a few straggling instances in even 
xinth century MSS. where «y appears to rhyme to e, the chief of 
which turn on apparently a dialectic pronunciation of saide as sede, 
which is also an orthography occasionally employed (p. 484, 1. 15, 
p. 481, 1. 33). Dr. Gill, 1621 {Logonomia p. 17), cites (sed) as a 
northern pronunciation for (said), and classes it with (saa) for (sai). 
Mr. Payne has pointed out similar cases in the Owl and Nightingale, v. 
349, 707, 835, 1779. The orthography sede occurs also, v. 472, 548, 
1293, and probably elsewhere.^ Mr. Payne also notes the less usual 
rhymes: bigrede upbreide 1411, misrede maide 1061, grede maide 
1335. These rhymes are certainly faulty, because in each case the 
ags. has a ^ in the second word but not in the first, and we cannot 
suppose them to have rhymed at this early period.^ In Floris and 

the action of an inserted coalescing i 
or e, according to the individual ten- 
dency of the language, passes into ai, 
or ei, or e and ie : prov. air, sp, aire 
from aer : prov. primairan (otherwise 
only 2J>'i»i6r prinner), port, primeiro, 
span, primero, it. primiero, from pri- 
ma rius ; prov. esclairar from esclariar 
which also exists ; prov. bais, port. 
beijo. span, beso from basium ; prov. 
fait, port, feito, span, hecho from f actus 
e being- palatalised into i. ... This 
vowel has suffered most in French, 
where its pure sound is often obscured 
into ai, e and ie. We must first put 
aside the common romance process, 
just noticed, by which this obscuration 
is effected by an inserted i as in air, 
premier, baiser, fait." Translated from 
Biez, Gr. der rom. Spr. 2nd. ed. i. 135. 

1 The Jesus Coll. Oxf. MS. reads 
seyde in each case. 

2 The orthography and rhymes of 
the Owl and Nightingale as exhibited 
in the Cott. MS. Calig. A. ix., fol- 
lowed by Wright, in his edition for the 
Percy Society, 1843, are by no means 
immaculate. The MS. is certainly of 
the XIII th century, before the introduc- 
tion of ou for (uu), that is, before 1280 
or probably before the death of Henry 
III., 1272, (so that, as has been con- 
jectured on other grounds, Henry II. 
was the king whose death is alluded to 
in the poem), and is contained in the 
same volume with the elder text of 
Lajamon, though it is apparently not 
by the same scribe. Nor should I be 
inclined to think that the scribe was a 
Dorsetshire man, although the poem 
is usually ascribed to Nicholas de 
Guildford, of Portisham, Dorsetshire. 

The confusions of e i, o e, e a, recall 
the later scribe of Havelok. Dreim 21, 
cleine 301, are obvious scribal errors, 
corrected to drem clene in the Oxf. MS., 
and : crei 334, in Oxf. MS. crey, although 
put in to rhyme with dai, must be an 
error for cri. We have cases of omitted 
letters in : rise wse o3, wrste toberste 
121, wlite wte 439, for wise, versteif), 
wite. There are many suspicious 
rh}TQes, and the following are chiefly 
assonances: worse mershe 303, hei- 
sugge stubbe 505, worde forworthe 
547, igremet of-chamed 931, wise ire 
1027, oreve idorve 1151, flesche cwesse 
1385, flijste vicst 405, and, in addition 
to the ei, e rhymes cited in the text, 
we have: forbreideth nawedeth 1381, 
in Oxf. MS. ne awede\. As to the 
present pronunciation of ay, ey in 
Dorsetshire, the presumed home of the 
poet, Mr. Barnes gives us very precise 
information : " The diphthongs ai or 
ay, and ei or ey, the third close long 
sound [that is, which usually have the 
the sound of a in mate'], as in May, 
hay, maid, paid, rein, neighbour, prey, 
are sounded — like the Greek at, — the 
a or e, the first open sound, as a in 
father, and the i or y as ee, the first 
close sound. The author has marked 
th a of diphthongs so sounded with a 
circumflex : as may, hay, maid, paid, 
vain, naighbour, pray." Foenis of 
Rural Life, 2nd ed., p. 27. — That is, 
in Dorsetshire the sound (ai), which 
we have recognized as ancient, is still 
prevalent. This is a remarkable com- 
ment upon the false rhymes of the 
MSS. Stratmann's edition, 1868, is of 
no use for the present investigation, on 
account of its critical orthography. 

Chap. VII. § I. AI AY, EI EY, IN SEVEN MSS. 639 

Elancheflur, Lumby's ed. occurs the rhyme : muchelhede maide 51, 
which is similarly faulty.^ See also p. 473 and notes there. We 
have likewise seen in some faulty west midland MSS. belonging to 
the latter part of the xvth century, (supra p. 450, n. 2), that ej/ 
was regarded as equivalent to e. In the Towneley Mysteries we 
also find ay, ey, tending to rhyme either with a or e. In fact we 
have a right to suppose that in the xv th century, at least, the pro- 
nunciation of ey, ay as (ee) was gaining ground, for we could not 
otherwise account for the MSS. mentioned, for the adoption of the 
spelling in Scotch in 1500, p. 410, n. 3, and for the fact that Hart, 
— who from various other circumstances appears to have been a 
West Midland man — seemed to know absolutely no other pronun- 
ciation of «?/ than (ee) in 1551.^ We have thus direct evidence 
of the coexistence of (ee, ai) in the xvi th century, each perhaps 
limited in area, just as we have direct evidence of the present co- 
existence of both sounds in high Grerman (p. 238), and Dyak (p. 474, 
note, col. 2). Such changes do not generally affect a whole body 
of words suddenly. They begin with a few of them, concerning 
which a difference prevails for a very long while, then the area is 
extended, till perhaps the new sounds prevail. We have an in- 
stance of this in the present coexistence of the two sounds (a, u) 
for short u, p. 175 and notes. It is possible that although Gill in 
1621 was highly annoyed at maids being called (meedz) in place of 
(maidz) by gentlewomen of his day (supra, p. 91, 1. 8), this very 
pronunciation might have been the remnant of an old tradition, 
preserved by the three rhymes just cited from the xin th century 
to the present day, although this hypothesis is not so probable as 
that of scribal error. And if it were correct, it would by no means 

1 On consulting the Auchinleck MS. of the text in the Auch. MS. runs thus, 

text of Floris et Elancheflur, the diffi- v. 518 : 

culty vanishes. Lumby's edition of To the king that jhe hem nowt 

the Cam. MS. reads, v. 49 : biwreie 

>u art hire ilich of alle >inge, Where thourgh thai were fiker to 

feoth of femblau?2t and of m?«rninge, dethe. 

Of fairneffe <^;^<^ of muchelhede, The editor suggests biwrei^e, which 

Bute>u ert a man and heo a maide ; would not be a rhyme. The real read- 
where the both of the second line makes ?S^ is mamfestly to deye, arising, as 
the third line altogether suspiciously ^^- Murray suggests, from the com- 
like an insertion. The Auchinleck Tl^ T rV^'^'A^'n'^.f'Z'^ll^ 
MS., according to the transcription ^^^^ ^^,, ^^^ ^^f"- ^^^ ^^tt. MSS. 
kindly furnished me by Mr. Halkett, constantly spelled -ayl, and hence we 
the librarian of the Advocates Library, must not be offended with the rhyme, 

Edinburgh, reads, V. 53: ^^^'S^ ^^^^^^^ 79^'. ^^^ ^^^^^. ^.^^ 

J. -T ^ T r ■<^ i- cvidcutly au uncertain pronunciation 

pou art ihch here of alle >inge ^f ^^^^ /grange word. 

Of semblant and of mournu.g 2 This day (9 July, 1869) a work- 

But >ou art a man an^ jhe is a maide ^^^ ^^^^^ ^^^'^^^^ L^-^^ ^^ 

Pons >ewif to nonce faide. ^^^ ^^11^^ ^^^^.^^^^ (spii-s^K).^ Had 

Another bad rhyme in the Cam. MS. he any idea that others said (spes'^l/) ? 

IS V. 533. The facts in the text are perhaps partly 

Hele ihc wulle and nojiing wreie accounted for by the influence of the 

Ower beire cuwpaignie Scotch orthography and pronunciation, 

which in the Abbotsford Club edition referred to on p. 637, n. 1, 



Chap. VII. § 1. 

prove that the general pronunciation of ay in all words from ags. 
was not distinctly (ai) and that the (ce) pronunciation was not 
extremely rare. 

In a former investigation it was attempted to shew that Norman 
French ei, ai, had at least frecj^uently the same sound (ai), supr^ 
pp. 453-459. Mr. Payne on the contrary believes that the sound 
was always pure (ee), and that the Norman words were taken into 
English, spellings and all, retaining their old sounds. He then 
seems to conclude that all the English ay, ey, were also pronounced 
with pure (ee), and maintains that this view agrees with all the 
observed facts of the case (p. 582). Prof. Rapp also, as we shall see, 
lays down that Early English Orthography was Norman, and as he 
only recognizes (ee) or (ee) as the sound of Norman ai, of course 
he agrees practically with Mr. Payne. Modem habits have induced 
perhaps most readers to take the same view, which nothing but the 
positive evidence of the practice of the xvi th century could easily 
shake. ^ But it would seem strange if various scribes, writing by 
ear, and having the signs e, ee, ea, ie, at hand to express th© sound 
(ee), should persist in a certain number of words, in always using 
ey, ay, but never one of the four former signs, although the sounds 
were identical. This is quite opposed to all we know of cacogra- 
phists of all ages, and seems to be only explicable on the theory of 
a real difference of sound, more marked than that of (ee, ee). Nay, 
more, some occasional blunders of e for ey, etc., would not render 
this less strange to any one who knows by painful experience (and 
what author does not know it ?) that he does not invariably write 
the letters he intends, and does not invariably see his error or his 
printer's or transcriber's errors when he revises the work. The 
mistake of e for ey we might expect to be more frequent than that 
of ay for e. When the writer is not a cacographist, or common 
scribe, but a careful theoretical orthographer as Orrmin or Dan 
Michel, the absolute separation of the spellings e, ey becomes 
evidence. We cannot suppose that Dutchmen when they adopted 
pais called it anything but (pais), why then should we suppose Dan 
Michel, who constantly employs the spelling pais^ pronounced 

1 I was glad to learn lately from so adraynk]?, agray]?!, etc., anpayri, apar- 

distinguished an English scholar as 
Prof. H. Morley that he was always of 
opinion that ay, ey, were (ai) and not 

2 Mr. Morris's index to Dan Michel's 
Ayenbite refers to p. 261, as contain- 
ing pese for peace. I looked through 
that page without discovering any in- 
stance of pese, but I found in it 11 in- 
stances of pais, pays and 3 of paysihle. 
Thinking Dan Michel's usages impor- 
tant, I have extracted those words given 
in the index, which of course does not 
refer to the commonest ags. words of 
constant occurrence. This is the list, 
the completeness of which is not gua- 
ranteed, though probable: adreynt, 

ceyue]7, apayre^, asayd, asayled, atrayt, 
bargayn, batayle, baylif, baylyes, bay]?, 
contraye, cortays, cortaysie, couaitise, 
dayes, defayled, despayred, eyder either, 
&yx=air, eyTen = egffs, ejse=ease, faili, 
fayntise, fornayce, germayn, graynes, 
greyner, longaynes, maimes, maine = 
retinue, maister, may den, maystrie, 
meseyse, meyster, nejebores, nejen, or- 
dayni ordenliche, oreysonne, paye = 
please, payenes =pagans, pays, paysible, 
plait, playneres, playni, playty, por- 
uaye]), porueyonce praysy, quaynte, 
queayntese, queyntise, raymi, [ags. reo- 
mianhryman, to cry out,] strait, strayni, 
tuay, uileynie, uorlay, Vf2ija.=gain, 
wayt, weyuerindemen, yfayled, zaynt. 

Chap. VII. § 1. AT AY, EI EY, IN SEVEN MSS. 641 

otherwise? And when we see some Erench words in Chaucer 
always or generally spelled with e which had an ai in French, as : 
resoun 276, sesoun 348, pees 2929, plesant 138, ese 223, 2672, 
why should we not suppose that in these words the (ee) sound 
was general, but that in others, at least in England, the (ai) sound 
prevailed ? Kay more, when we find ese occasionally written eyu 
for the rhyme in Chaucer (supra p. 250 and note 1, and p. 265), 
as it is in Dan Michel's prose, why should we not suppose that two 
sounds were prevalent, just as our own (niidh'i, naidh'j) for neither, 
and that the poet took the sound which best suited him ? This 
appears to me to be the theory which best represents all the facts 
of the case. It is also the theory which best accords with the 
existing diversities of pronunciation within very narrow limits in the 
English provinces. It remains to be seen how it is borne out by the 
orthography of the Ha. Harleian 7334, and the six newly published 
MS. texts, E. Ellesmere, He. Hengwrt, Ca. Cambridge, Co. Corpus, 
P. Petworth, and L. Lansdowne of the Canterbury Tales. Eor this 
purpose I have looked over the prologue and Knightes Tale, and 
examined a large number, probably the great majority of the cases, 
with the following results. The initial italic words, by which the 
lists are arranged, are in modem spelling, and where they are 
absent the words are obsolete. Where no initials are put, all the 
MSS. unnamed agree in the preceding spelling so far as having one 
of the combinations ai, ay, ei, ey is concerned, small deviations in 
other respects are not noted, but if any other letter is used for one 
of the above four it is named. The numbers refer to the lines of 
the Six Text edition, and they have frequently to be increased 
by 2 for Wright's edition of the Harleian MS. 

List of Words coNTAiNma AY, EY in the Prologue and Knightes Tale. 

Anglosaxon and Scandinavian maidens, maydens 2300 

Words. nails, nayles 2141 

neighbour, nyjhebour Ca., neighebore 
again, agayn 991 535 

against, ajens Ca., ageyns 1787 neither, neither 1135 

aileth, eyleth 1081 nigh, neigh H. He., neyh Co., nyghe 
ashes, aisshes Co., asshen 2957 P., nyhe L., nyh Ca„ ny E., 732 

bewray, bewreye 2229 said, seyde 219, 1356, and frequently 

day, day, 19 and frequently say, seyn 1463 

die, deyen Ca., Co., dyen E. He. P. seen, seyn E. He. Ca. Co. L., seen Ha., 

dyjen L. 1109, deyde 2846 sene P. 2840 

dry, dreye Ca., drye 420, 1362, dreye slain, slayn 992, 2038, 2552, 2708 ; 

[rh. weye] 3024 slayn P. L., sleen 1556, sle sleen 

dyer, deyer Ha., dyere 362 1859 

eye, eye E. Ca., eyghe P., yhe Ha. L., sleight, sleight 604 

iye He. 10, eyen E. He., eyghen spreynd Ha. E. He. Co. P., sprend Ca., 

Ha. P., eyjyyn Ca., yghen Co., spriued L. 2169 

yhen L. 267 and frequently two, tweye 704 

fain, fayn 2437 waileth, wayleth 1221 

fair, faire 1685. 1941 way, way 34, 1264, and often. 

flesh, fleissh Ha. Co., flessh 147 weighed, weigheden 454 

height, heght P., heighte 1890 whether, wheither E. He., whethir Ha., 
laid, leyde 1384 and frequently whe))er Ca. Co. L., whedere P., 

lay, lay 20 and frequently 1857 



Chap. VII. § 1. 

French Words. 

acquaintance, aqueyntaunoe 245 
ciieul, aid E. He. Ca. ayel Ha., ayell 

Co. L. eile P. 2477 
aivy eir 1246 
apayd [rh. ysaid] 1868 
apparelling, apparaillyngo 2913 
array, array 41 73, and often. 
attain, atteyne 1243 
avuilcth, auailletli 3040 
bargains, bargaynes 282 
barren, barayne 1244, baran L., bareyn 

battle, bataille 988, 2540 
braided, breided P., broyded E. He. 

Ca. Co., browded Ha. L, 1049 
caitiff, catiflf P., caytyf 1552, 1717, 1946 
certain, certe)ii 204 and often. 
chain, chepie 2988 
chutaigne, chastc}Ti 2922 
chieftain, cbevetan Ha., chieftayn 2555 
company, compaignye E. He. Co. P., 

cumpanye Ca., companye Ha. L. 

331, compaignye E. He. L., cum- 
panye Ca. Co. P., company Ha. 

2105, 2411 
complain, compleyn 908 
conveyed, conuoyed E., conveyed 2737 
counsel, conseil Ha. E, He. Co. P., 

counsel L., cuntre Ca. 3096 
courtesy, curteisie E. He. Ca., curtesie 

Ha. Co. P. L. 46, 132 
dais, deys Ha. E. He. Ca. Co. P. dese 

[rh. burgeise] L. 370 
darreyne, 1609, 2097 
debonnair, debonnaire [rh. faire] 2282 
despair, dispeir 1245 
dice, deys Ca., dys 1238 
disdain, disdejTi 789 
diijplayeth, desplayeth 966 
distraineth, destreyneth 1455, 1816 
dozen, doseyoie 578 
fail, faille 1854, 2798 
finest, feynest Ca., fynest 194 
florin, floreyn Ca. Co. P., floren Ha. 

L., floryn E. He. 2088 
franklins, frankeleyns 216 
fresh, fresshe Ha. E. He. P. L., frossche 

Ca., freissche Co., 92, [freischHa.] 

2176, 2622 
furnace, forneys 202, 559 
gaineth, gayneth 1176, 2755 

gay, gay 73 

golyardeys 560 

harnessed, harneysed 114, 1006, 1634, 

kerchiefs, keverchefs Ha., couereheis 
Ca. [the proper Norman plural, 
according to Mr. Payne], couer- 
chiefs E. He. Co. L., couerchefes 
P. 453 

leisure, leyser 1188 
Magdalen, Maudelayne 410 
maintain, raaynte)iie H. E., ma}Titene 

He. Ca. Co. P., maiten L. 1778 
master, mystir Ca., maister 261 
mastery, maistrie 165 
mejTied 2170 
money, moneye 703 
ordained, ordeyncd 2553 
paid, ypayed 1802 
pain-ed, peyned 139, pejTie 1133 
painted, pcyntid 1934, 1975 
palace, paleys 2513 
palfrey, palfrey 207, 2495 
plain, pleyn 790, 1464 
plein, pleyn 315 
portraiture, portreiture Ha. E. He. Ca. 

Co., pourtrature P. L. 1968, [pur- 

treture Ha.] 2036 
portray, portray 96 
portrayer, portreyor Ha., portreitour 

E., purtreyour He., purtreiour 

Co., purtraiour P., portretour Ca., 

purtreoure L., 1899 
portraying, portraying Ha., portreying 

Ca. Co.. purtraiynge P., por- 

treyynge E. He., purtreinge L. 

pray, prey en 1260 
prayer, prayer 2226 
purveyance, purveiance E. He., pur- 

ueance Ha. Co. P. L. puruyance 

Ca. 1665, purueiance E. H., pur- 

ueance Ha. Co. P. L., puruyance 

Ca. 3011 
quaint 1531, 2321, 2333, 2334 
raineth, reynith 1535 
reins, reynes 904 
sovereign, souereyn 1974 
straight, streite 457, stryt Ca., streyt 

suddenly, sodanly L., sodeynly 1530, 

sodeinliche 1575 
sustain, susteyne Ca. L., sustene 1993 
trace, trays 2141 
turkish, turkeys 2895 
turneiynge E. He. Co. turneynge Ha., 

turnyinge Ca. tornynge L., tor- 

namente P. 2557 
vain, veyn 1094 
vasselage Ha. E. He. Co. L., vassalage 

P., wasseyllage Ca. 3054 
vein, veyne 3, 2747 
verily, verraily E. He. Ca. Co. verrely 

P. L., verrily Ha. 1174. 
very, verray 422 
villany, vileynye E. He., velany Ca., 

L., vilonye Ha. Co. P. 70, [vilanye 

Ha.] 740 
waiting, waytinge 929 

Chap. VII. § 1. 



The general unanimity of these seven MSS. is certainly remarkable. 
It seems almost enough to lead the reader to suppose that when 
he finds the usual ay, ey replaced by «, e, i in any other MSS., the 
scribe has accidentally omitted one of the letters of the diphthong, 
which being supplied converts a, e, i into ay, ey, ai or ei respectively. 
Thus when in v. 1530 all but L. use ey or ay, and in v. 1575 all, in- 
cluding L., use ey in sodeynly, sodeynliche, we cannot but conclude 
that sodanly in L. 1530, is a clerical error for sodaynly. We have 
certainly no right to conclude that the a was designed to indicate 
a peculiar pronunciation of a as ay or conversely. But it will be 
best to consider the variants seriatim as they are not many in 

Consideration of Variants in the Last List. 

Anglosaxon and Scandinavian 

Against 1787 has still two sounds 
(^gemst-, "Bgenst-) which seem to cor- 
respond to two such original sounds as 
(again* agen*). 

Ashes, aisshes Co. 2957 represented 
really a duplicate form, as appears from 
its having been preserved into the 
XVI th century, p. 120, 1. 6. 

Die 1109, see variants on p. 284. 

Dry 420, see variants on p. 285. 

Dyer, the general orthography dyer 
362 is curious, for the ags. deagan 
would naturally give deyer, which how- 
ever is only preserved in Ha., the rest 
giving dyere, and the Promptorium 
having dyyn; Ha. has deye in 11037. 
It would almost seem as if habit had 
confused the two words dye, die, and 
hence given the first the same double 
sound as the second. There is no 
room for supposing the sound (dee) in 
either case. 

Eye 10, see variants on p. 285. 

Flesh, 147 is one of the words men- 
tioned on p. 265, as having two spell- 
ings in Ha. see also p. 473 note 1, for 
a possible origin of the double pronun- 

Height, heght P. 1890 is of course 
a clerical error for heighte. 

Neighbour 535, follows nigh in its 

Nigh 732, 535. The variants here 
seem to shew that this word should be 
added to the list given on pp. 284-6, 
as having a double pronunciation, 
especially as we have seen that the (ii) 
sound is preserved in Devon, p. 291, 
as it is in Lonsdale. 

Seen. The orthography seyn 2840 
for seen is supported by too many 
MSS. to be an error, it must be a du- 

plicate form, retaining in the infinitive 
the expression of the lost guttural, 
which crops up so often in different 
parts of this verb, Gothic saihivan, 
compare the forms on p. 279. 

Slay 992, see p. 265 ; the double 
sound (ee, ai) may have arisen from the 
double ags. form, without and with the 
guttural, the latter being represented 
by (ai) and the former by (ee), which 
is more common. 

Sprei)id, isprend, isprind 2169 must 
be merely clerical errors for ispreined, 
as in most MSS., because both words 
rhyme with ymeynd, which retains its 
orthography in each case. 

Whether, 1857, has certainly no 
more title to (ai) than heat or them, 
but nevertheless we have seen Orrmin 
introduce the (i) or (j) into these words, 
p. 489, hence it is not impossible that 
there may have been some provincials 
who said ivheider, but still it is more 
probable that the ei of E. and He. in 
1857 are clerical errors. The word is 
not common and I have not noted 
another example of it in E. He. 

French Words. 

Barren, baran L. 1977, must be a 
clerical error for barayn. 

Braid 1049, seems to have had 
various sounds, corresponding to the 
ags. bregdan, icel. bregda, and to the 
French broder, which would give the 
forms breyde, browde. while broyde 
would seem to be an uncertain, or mis- 
taken mixture of the two (braid-e, 
bruud'e, bruid'e). We do not find 
brede (breed-e). but as the g was some- 
times omitted even in ags. it would 
have been less curious than brayde. 

Caitiff. The orthography catiff P. 
1552, 1717, 1946, being repeated in 



Chap. VII. $ 1. 

three places, although opposed to the 
other six MSS. which dotermine caytif 
to be the usual form, may imply a dif- 
ferent pronunciation rather than be a 
clerical error. The French forms of 
this derivative of the Latin captivus, 
as given by Roquefort are very numer- 
ous, but all of them contain i, or an e 
derived from ai, thus: caitif, caiptif, 
caitieu, caitis, caitiu, caitivie, cetif, 
cetis, chaitieu, chaitif, chaitis, chaitiu, 
cheitif, chetif, chety, quaitif, quetif. 
Roquefort gives as Provencal and 
Languedoc forms : caitiou, caitious, 
caitius, caitivo. The Spanish cautivo 
has introduced the labial instead of the 
palatal modification, while the Italian 
only has preserved the a pure by as- 
similating Pf thus, cattivo. If then 
the a in P. was intentional, it was very 

Chieftain^ cheveten Ha. 2555, should 
according to the general analogy of 
such terminations be cheveteyn, and it 
will then agree with the other MSS. 

Company. In compaignye 331, 2105, 
2411, the i is conceived by M. Fran- 
cisque Michel to have been merely 
orthographical in French, introduced 
to make gn moicille, just as i was intro- 
duced before II to make it mouille. 
Compare also p. 309, n. 1, at end. It 
is very possible that both pronuncia- 
tions prevailed (kumpainu'*e, kum- 
panere) and that the first was con- 
sidered as French, the latter as Eng- 
lish. There is no room for supposing 
such a pronunciation as (kumpeenu'e) 
with (ee). 

Conveyed. Conuoyed E. 2737 is not 
a variant of the usual conueyed., but 
another word altogether, a correction 
of the scribes. 

Counsel^ counsel L. 3096, is probably 
a clerical error for counseil as in the 
other MSS. 

Courtesy. Curteisye 46, vileynye 70, 
may be considered together. They 
were common words, and the second 
syllable was usually unaccented, where- 
as in curteis, vileyn^ it was frequently 
accented. Hence we cannot be sur- 
prised at finding ey strictly preserved 
in the latter, but occasional deviations 
into non-diphthongal sounds occurring 
in the former. Careful scribes or 
speakers seem, however, to have pre- 
served the ey of the primitive in the 
derivative. The vilonye of Ha. Co. P. 
70, which is replaced by vilanye in Ha. 

740, serves to corroborate this view, 
as evidently the scribe did not know 
how to wTite the indistinct sound he 
heard, a difficulty well known to all 
who have attempted to write down 
living sounds. See also Mr. Payne's 
remarks, supra p. 585. To the same 
category belong the variants of por- 
traiture, purveyance, verily. 

Dais, dese L. for deys=did\% 370, in 
opposition to the six other MS. is pro- 
bably a clerical error for deyse the final 
e being added also to the rhyming 
word burgeise in L. which retains the i. 

Bice. Deys Ca. 1238 for dys is 
clearly an error as shewn by the rhym- 
ing word paradys, but dys itself seems 
to have been accommodated to the 
rhyme for dees, which occurs in Ha. 
13882, and is the natural representa- 
tive of the French des. 

Finest. The orthography fey nest 
Ca. 194, must be a clerical error. 

Florin. The floren, florin, floreyn 
2088 may be concurrent forms of a 
strange word, and the last seems more 
likely to have been erroneous. 

Fresh 92, had no doubt regularly 
(ee), but the older (ai) seems to have 
been usual to some, the frosshe of Ca. 
is a provincialism of the order noted 
on p. 476 

Kerchiefs. Couercheis Ca. 453, is 
probably a mere clerical error for 
couerchefs, i having been written for 
/, as we can hardly suppose the provin- 
cial scribe of Ca., to have selected a 
Norman form by design. 

Maintain. Maynteyne 1778, sus- 
te)Tie 1993, belong to the series of words 
derived from tenere. There is no dis- 
agreement respecting the ay in the 
first syllable of maynteyne ; sustene is 
fully supported by the rhyme, p. 265, 
1. 1, and hence mayntene, sustene are 
probably the proper forms. I have 
unfortunately no note of the Chau- 
cerian forms of obtain, detain, retain, 
contain, appertain, entertain, abstain, 
but probably -tene would be found the 
right form. The spelling ey and pro- 
nunciation (ai) may have crept in 
through a confusion with the form 
-teyne =:Ju2it. -tingere, of which I have 
also accidentally been guilty p. 265, 
1. 25, as : atteyne, bareyne, must rhyme, 
1243, 8323, and as -stringere produces 
-streyne 1455, 1816 in all MSS. 

Master, mystir Ca. 261 for master is 
probably a clerical error. 

Chap. VII. § 1. 



Fortraiture 1968, portrayer 1899 ; 
the variants may be explained as in 
Courtesy, which see. 

Fortraying. In portreyyng, por- 
treyng 1938 there is an omission of 
one y on account of the inconvenience 
of the yy in the first form, overcome 
by changing the first y into i in P. 

Furveyance 1165, the variants may 
be explained as in Courtesy, which see. 

Straight. Stryt Ca. 1984, must be 
a clerical error for streyt, as the ab- 
sence of e is quite unaccountable. 

Suddenly. Sodanly L. 1530 must, as 
we have seen p. 643, be an error for 

The natural effect of tliis examination lias been to place the 
variants rather than the constants strongly before the reader's mind. 
He must therefore recollect that out of the total of 111 words the 
following 73, many of which occur very frequently, are invariably 
spelt with one of the phonetically identical forms ai^ ay, ei, ey., 
in each of the seven MSS. every time they occur : — 

again, aileth, bewray, day, fain, fair, dozen, fail, franklins franJceleyns, fur- 
laid, lay, maidens, nails, neither, said, 
say, sleight, two tweye, waileth, 
way, weighed. acquaintance, dieul^ 

Sustain 1993 see Maintain. 

Turneynge Ha. 2557 ; the variants 
are to be explained as those oi portray- 
ing, which see. 

Verily 1174, the variants may be 
explained as in Courtesy, which see. 

Villany 70, see Courtesy. 

WasseyUage Ca. 3054, certainly 
arose from a confusion in the scribe's 
mind, vasselage valour being unusual, 
he reverted to the usual wasseyl for an 
explanation, and in wasseyl we have an 
ey for an ags. ce, which may be com- 
pared with ey for ea in Orrmin, supra 
p. 489. 

air, apayd, apparelling apparaillynge, 
array, attain, availeth, bargains, battle 
bataille, certain, chain, chataigne, com- 
plain, darreyne, debonnair, despair, 
dice, disdain, displayeth, distraineth, 

nace forneys, gaineth, gay, golyardeysj 
harnessed harneysed, leisure, Magdalen 
Maudelayne, mastery, meyned, money, 
ordained, paid, pained, painted, palace 
paleys, palfrey, plain, plein, portray, 
pray, prayer, quaint, raineth, reins, 
sovereign, trace trays, turkish turkeys, 
vain, vein, very, wailing. 

On the other hand, the variants only affect 38 words, of which 
few, except those already recognized to have two forms in 
use, occur more than once, while the variants confined to one or 
two MSS. display no manner of rule or order, and are far from 
shewing a decided e form as the substitute for ay, ey. They may 
be classified as follows : 

15 Clerical Errors : height 
heght, spreyned sprend sprined, whether 

wheither, barren baran, chieftain, 

chevetan, counsel counsel, dice deys, 
finest feynest, kerchiefs couercheis, 
maintain maynteyne mayntene, master 
mystir, straight stryt, suddenly sodanly, 
sustain susteyne, turneiynge turnyinge 

12 Double Forms : ashes aisshes 
asshen, die deyen dyen, dry dreye drye, 
dyer dyere deyer, eye eighe yhe, flesh 
fleissh flessh, neighbour neighebore 
nyjhebour, nigh neigh nyghe, seen seyn 

seen, slain slayn sleen, braided 

breided browdid, fresh fresshe freisshe. 

6 Indistinct Unaccented Sylla- 

bles : courtesy courteisie curtesie, por- 
traiture portreiture pourtrature, por- 
trayer portreyor purtreoure, purvey- 
ance purveiance purueance puruyance, 
verily verraily verrely verrily, villany 
vileynye velany vilonye. 

5 Miscellaneous : m^Y^^ may have 
been occasionally catiffd& well as caytif 

conuoyed was a difi'erent reading, 

not an error for conveyed florin 

being a foreign coin may have been 
occasionally mispronounced floreyn, 

portreing was an orthographical 

abbreviation of portreiynge was- 

seyllage was a manifest error for the 
unusual vasselage, the usual wasseyl oc- 
curring to the scribe. 

The variants, therefore, furnish almost as convincing a proof as 
the constants, that ay, ey represented some sound distinct from e 

646 TREATMENT OF FINAL E. Chap. VII. $ 1. 

(ee). ^ But if there was a distinct sound attachable to these com- 
binations ay, ey, in Chaucer's time, what could it have possibly been 
but that (ai) sound, which as we know by direct evidence, subsisted 
in the pronunciation of learned men and courtiers (Sir T. Smith was 
secretaiy of state) during the xvi th century, and which the spelling 
used, and no other, was calculated to express, and was apparently 
gradually introduced to express. The inference is therefore, that 
Chaucer's scribes pronounced ay, ey as (ai) and not as (ee), and 
where they wished to signify the sound of (ee), in certain well- 
known and common I^orman words, they rejected the Norman or- 
thography and introduced the truly English spelling e. The in- 
ference again from this result is that there was a traditional English 
pronunciation of I^orman ai, ei, as (ai), which may have lasted long 
after the custom had died out in ^N'ormandy, on the principle already 
adduced (p. 20), that emigrants preserve an older pronunciation. 

Treatment of Final E in the Critical Text. 

As the following text of the Prologue is intended solely for the 
use of students, it has been accommodated to their wants in various 
ways. First the question of final e demanded strict investigation. 
The helplessness of scribes during the period that it was dying out 
of use in the South, and had already died out in the :N'orth, makes 
the new MSS. of little value for its determination, the Cambridge 
and Lansdowne being evidently written by JS'orthern scribes to 
whom a final e had become little more than a picturesque addition. 
It was necessary therefore to examine every word in connection 
with its etymology, constructional use, and metrical value. In 
every case where theory would require the use of a final e, or other 
elided letter, but the metre requires its elision, it has been replaced 
by an apostrophe. The results on p. 341 were deduced from the 
text adopted before it had been revised by help of the Six-Text 
Edition, and therefore the numbers there given will be slightly 
erroneous ^ but the reader will by this means understand at a glance 
the bearing of the rules on p. 342. 

The treatment of the verbal termination -ede, required particular 
attention. There are many cases in which, coming before a con- 
sonant, it might be -ed' or -'de, and it was natural to think that the 
latter should be chosen, because in the contracted forms of two 
syllables, we practically find this form ; thus : fedde 146, bledde 
145, wente 255, wiste 280, spente 300, coude 326, 346, 383, kepte 
442, dide 451, couthe 467, tawghte 497, cawghte 498, kepte 512, 
wolde 536, mighte 585, scholde 648, seyde 695, moste 712 and 

1 The number of elisions of essential lowing are examples: palmer's 13, 

^, stated at 13 on p. 341, has been re- servawnt's 101, fether's 107, finger's 

duced. The only important one left is 129, hunter's 178, grayhound's 190, 

meer' 541, and that is doubtful on ac- sleev's 193, tavern's 240, haven's 407, 

count of the double form of the rhym- housbond's 460, aventur's 795. Of 

mg word milleer. see p. 389. The course (') is not used as the mark of 

number of plural -es treated as -s has the genitive cases, but only to shew a 

been somewhat increased. The fol- real elision. 

Chap. VII. § 1. TREATMENT OF FINAL E. 647 

many others. Eut even here it is occasionally elided. Mr. Morris 
observes that in the Cambridge MS. of Eoethius, and in the elder 
Wycliffite Yersion (see below § 3), the -ede is very regularly written. 
This however does not prove that the final e was pronounced, be- 
cause the orthography hire^ here, oure, youre, is uniform, and the 
elision of the final -e almost as uniform. The final e in -ede might 
therefore have been written, and never or rarely pronounced. It is 
certain that the first e is sometimes elided, when the second also 
vanishes, as before a vowel or A in: lov'd' 206, 533, gam'd' 534, etc. 
But it is also certain that -ed'' was pronounced in many cases with- 
out the (9, supra p. 355, art. 53, Ex. Throughout the prologue I 
have not found one instance in which -ede, or -de, was necessary to 
the metre, ^ but there are several in which -ed\ before a vowel, is 
necessary. If we add to this, that in point of fact -ed"* remained in 
the XVI th century, and has scarcely yet died out of our biblical 
pronunciation, the presumption in favour of -ed^ is very strong.^ On 
adopting this orthography, I have not found a single case in the 
prologue where it failed, but possibly such cases occur elsewhere, 
and if so, they must be compared to the rare use of hadde, and 
still rarer use of were, here for the ordinary hadd\ wer\ her\ 

The infinitive -e is perhaps occasionally lost. It is only saved 
by a trisyllabic measure in : yeve penawnce 223. If it is not 
elided in help'' 259, then we must read whelpe 258, with most MSS. 
but unhistorically. On the other hand the subjunctive -e remains 
as : ruste 500, take 503, were 582, spede 769, quyte 770. 

Medial elisions must have been common, and are fully borne out 
by the Cuckoo Song, p. 423. Such elisions are: ev'ry 15, 327, 
ev'ne 83, ov'ral 249, ov'rest 290, rem'nawnt 724, and : mon'th 92, 
tak'th 789, com'th 839. The terminations -er, -el, -en, when run 
on to the following vowel, should also probably be treated as 
elisions. As respects -er, -re, I have sometimes hesitated whether to 
consider the termination as French -re, or as assimilated into English, 
under the form -er, but I believe the last is the right view, and in 
that case such elisions as: ord'r he 214, are precisely similar to : 
ev'ry 15, and occasion no difficulty. Similarly, -el, -le, are both 
found in MSS., but I have adopted -el, as more consonant with the 
treatment of strictly English words, and regarded the cases in which 
the I is run on to the following word, as elisions, thus : simp'l and 
119. Such elisions are common in modem English, and in the case 
of -le, they form the rule when syllables are added, supra p. 52. 
In : to fest'n' his hood 195, we have an elision of e in en, and a final 
e elided, the full gerundial form being to festene, as it would be 
written in prose. 

1 The plural weygheden 454, is not tahlys^ sadlys^ fadrys^ modrys^ but its 
in point. subsequent restoration, accompanied 

2 Mr. Murray observes that lovde by a suppression of the y before the s, 
would be an older form than loved for in the more recent forms tabylls 
lovede^ and grounds his observation on sadylls^ fadyrs^ modyrs. These analo- 
the fact of the similar suppression gies are valuable. All that is implied 
of the y before I in tabyll^ sadyll, in the text is that the form -ed seems 
fadyr, modyr, in the old Scotch plurals to have prevailed in Chaucer. 

648 Chaucer's metre. Chap. VII. § i. 

As the text now stands there is no instance of an open c, that is, 
of final e preservx'd before a vowel (supra p. 341, 1. 2. p. 363, art. 
82, and infra note on v. 429), but there is one instance of final e 
preserved before he^ (infra note on v. 386). 

Metbical Peculiarities of Chaucer. 

The second point to which particular attention is paid in this 
text is the metre. Pains have been taken to choose such a text as 
would preserve the rhythm without violating the laws of final ^, and 
without having recourse to modem conjecture. For this purpose 
a considerable number of trisyllabic measures (supra p. 334) have 
been admitted, and their occurrence is pointed out by the sign iii 
in the margin. The 69 examples noted may be classified thus : 

t- , arising from the running on of i to a following vowel, either in two 
words as : many a 60, 212, 229, etc., bisy a 321, carl' a 130, studi' 
and 184, or in the same word, as : luvieer 80, curious 196, bisier 321, 
which may be considered the rule in modern poetry, see 60, 80, 130, 
184, 196, 212, 229, 303, 321, 322, 349, 350, 396, 438, 464, 630, 

560, 764, 782, 840, instances 20 

-tfr, arising from running this unaccented syllable on to a following 

vowel, in cases where the assumption and pronunciation of -V would 

be harsh, as : deliver, and 84, sommer hadd' 394, water he 400 ; and 

in the middle of a word, as : colerik 587, leccherous 626 ; instances 5 

-ely not before a preceding vowel, as : mesurabel was 435, mawncipel 

was 567, mawncipel sett' 586, instances 3 

-m, not before a preceding vowel, as : yeomen from 77 ; or before a pre- 
ceding vowel or A, where the elision 'w would be harsh, as : writen 

a 161, geten him 291, instances 3 

-e, arising from the pronunciation of final ^, where it seems unnecessary, or 
harsh, to assume its suppression, as 88, 123, 132, 136, 197, 208, 223, 
224, 276, 320, 341, 343, 451, 454, 475, 507, 510, 524, 537, 550, 630, 

648, 650, 706, 777, 792, 806, 834, 853, instances 29 

Miscellaneous, in the following lines, where the trisyllabic measures are 
italicised for convenience. 

Of Engelond', to Cawnterber?/ thetj tvende. 16 ^ 

To CaAvnterbery withful devout corage. 22 

His heed was balk*^, and schoon as any glas. 198 

And thryes hadd' she been at Jerusalem. 463 

"Wyd was his ^oxisch and houses fer asonder. 491 ^ instances 

He was a schei^perd, and not a mercenarie. 514 

He waited after no pomp' and reverence. 525 

Ther coude no man bring' him in arrerage. 602 

And also war' him of a significavit. 662 

Total 69 

It would have been easy in many cases by elisions or slight 
changes to have avoided these trisyllabic measures, but after con- 
sidering each case carefully, and comparing the different manu- 
scripts, there did not appear to be any sufficient ground for so doing. 

Allied to trisyllabic measures are the lines containing a super- 
fluous unaccented syllable at the end, but to this point, which was 
a matter of importance in old Italian and Spanish versification, and 
has become a matter of stringent rule in classical French poetiy, no 
attention seems to have been paid by older writers, whether French 
or English, and Chaucer is in this respect as free as Shakspere. 

Chap. VII. § 1. CHAUCER's METRE. 649 

There are a few cases of two superfluous unaccented syllables, com- 
parable to the Italian versi sdruccioU, and these have been indicated 
by (-{-) in the margin. There are only 6 instances : berye merye 
207, 208, apotecaryes letuaryes 425, 426, miscarye mercenarye 513, 
514, all of which belong to the class i-, so that the two syllables 
practically strike the ear as one. 

But there are also real Alexandrines, or lines of six measures, 
which do not appear to have been previously noticed, and which I 
have been very loth to admit. These are marked vi in the margin. 
There are four instances. In : 

But sore wepte sche if oon of hem wer' deed. 148 

the perfect unanimity of the MSS., and the harsh and unusual 
elision of the adverbial -e in sore, and the not common elision of the 
imperfect e in wepte, which would be necessary to reduce the line to 
one of five measures, render the acceptance of an Alexandrine im- 
perative, and certainly it is effective in expressing the feeling of 
the Prioresse. In : 

Men mote yeve silver to the pore freres. 232 

the Alexandrine is not pure because the caesura does not fall after 
the third measure. But the MSS. are unanimous, the elisions fnof 
yev' undesirable, and the lengthening out of the line with the tag 
of ''the pore freres," seems to indicate the very whine of the 
begging friar. In 

"With a thredbare cop', as a pore scoleer. 260 

the pore which lengthens the line out in all MSS., seems introduced 
for a similar purpose. The last instance 

I ne sawgh not this yeer so mery a companye. 764 

is conjectural, since no MS. gives the reading complete, but: I ne 
sawgh, or : I sawgh not, are both unmetrical, and by using both 
we obtain a passable Alexandrine, which may be taken for what it 
is worth, because no MS. reading can be accepted. 

The defective first measures to which attention was directed by 
Mr. Skeat, supra p. 333, have been noted by ( — ), and a careful 
consideration of the MSS. induces me to accept 13 instances, 1, 76, 
131, 170, 247, 271, 294, 371, 391, 417, 429, 733, 778, though 
they are not all satisfactory, as several of them (131, 247, 271, 
391, 778) offend against the principle of having a strong accent on 
the first syllable, and two (417, 429) throw the emphasis in rather 
an unusual manner, as : weel coud' he, weel knew he, where : weel 
coud^ he, well knew he, would have rather been expected, but there 
is no MS. authority for improving them. 

Three instances have been noted of saynt forming a dissyllable, 
as already suggested, (supra pp. 264, 476), one of which (697), 
might be escaped by assuming a bad instance of a defective first 
measure, but the other two (120, 509,) seem clearly indicated 
by MS. authority. See the notes on these passages. They are 
indicated by ai in the margin.^ 

1 Mr. Murray has observed cases in then it had its Scotch value (a^), supra 
Scotch in vsrhich ai was dissyllabic, but p. 637, n. 1. He cites from Wyn- 


650 Chaucer's French words. Chap. VII. § i. 

Chaucer's Treatment of French Words. 

The third point to which attention is directed in printing the 
text of the prologue, is linguistic rather than phonetic, but seemed 
of sufficient interest to introduce in a work intended for the use of 
the Chaucer Society, namely, the amount of French which Chaucer 
admitted into his English. ''Thank God! I may now, if 1 like, 
turn Protestant!" exclaims Moore's Irish Gentleman on the evening 
of 16th April, 1829, when the news of the royal assent to the 
Catholic llelief Bill reached Dublin.^ And in the same way it 
would appear that the removal of the blockade on the English 
language, when after "j^e furste moreyn," 1348, ''John Comwal, 
a maysterc of grammere, chaungede J?e lore in gramere scole,"^ and 
Edward III. enacted in the 36th year of his reign, 1362-3, that all 
pleas should be pleaded and judged in the English tongue, the 
jealous exclusion of Erench terms from English works, which marks 
the fomier period, seemed to cease, and English having become the 
victor did not disdain to make free use of the more " gentle" 
tongue, in which so many treasures of literature were locked up. 
Even our older poems are more or less translations from the French, 
though couched in unmistakable English. But in the xrvth 
century we have Gower writing long poems in both languages, 
and Chaucer familiar with both, and often seeking his originals in 
French. The people for whom he principally wrote must have 
been also more or less familiar with the tongue of the nobles, and 
large numbers of French words must have passed into common use 
among Englishmen, before they could have assumed English in- 
flectional terminations. We have numerous instances of this in 
Chaucer. Whenever a French verb was employed, the French 
termination was rejected, and an English inflectional system sub- 
stituted. Thus using italics for the French part, we have in the 
prologue: perced 2, engendered 4, 421, inspired 6, esed 29, honoured 
50, emhrouded 89, horneysed 114, entuned 123, peyned 139, ros^edl47, 
jpincJiQd 151, gawded 159, croimed 161, purfyled 193, farsed 233, 
accorded 244, enryned 342, chaunged 348, passed 464, encomhred 
508, spyced 526, jpuiiish'd 657, trussed 681, feyned 705, assembled 
717, served 749, grawnted 810, pray^den 811, reuled 816, studieih. 

841. floiitm^^ 91, harping^ 266, o^Hng' 450, 489, aswyling 661, 

cry^ 636, rosf, hroyW , frye 383, rehers'' 732, feyne 736. Again 

we have an English adjective or adverbial termination affixed to 
French words, as: specially 15, fetislj 124, 273, certainly 235, 
solemnelj 274, staatly 281, estaat]lch 140, verrayly 338, really 

town's Orygynal Cronykil of Scotland, search of a religion, by Thomas Moore, 

circa 1419-30, in reference to Malcolm chap. i. 


Malcolm kyng, be Liwchful get, ^ ggg i\^q whole noteworthy passage 

Had on his wyf Saynt Margret. from Trenisa's translation of Higden, 

Where, however, Margret might rather printed from the Cott. MS. Tiberius 

have been trissyllabic. D. VII., by Mr. E. Morris, in his 

Specimens of Early English, 1867, 

* Travels of an Irish gentleman in p. 339. 

Chap. VII. § 1. CHAUCER's FRENCH WORDS. 651 

prively 652, playnlj 727, properly 729, rudely 734. deWle^s 

582. In esy 441, pomely 616, we have rather the change of the 

French -e into -y, which subsequently became general, but the ese 
remains in : esely 469. In : daggeQV 113, 392, we have a substan- 
tive with an English termination to a French root, ^ootmantel 
472, is compounded of an English and French word. In : didiVLawnce 
211, loodmann^y^ 403, deyery^ 577, French terminations only are 
assumed. A language must have long been in familiar use to 
admit of such treatment as this. What then more Kkely than the 
introduction of complete words, which did not require to have their 
terminations changed ? The modern cookery book and fashion 
magazines are full of French words introduced bodily for a similar 
reason. Of course the subject matter and the audience greatly 
influence the choice of words, and we find Chaucer sensibly changing 
his manner with his matter — see the quantity of unmixed English 
in the characters of the Yeman, the Ploughman, and the Miller. 
To make this admixture of French and English evident to the eye, 
all words or parts of words which may be fairly attributed to French 
influence, including proper names, have been italicised, but some 
older Latin words of ecclesiastical origin and older JN^orman words 
have not been marked and purely Latin words have been put in 
small capitals.^ The result could then be subjected to a numerical 
test, and comes out as follows : 

Lines containing no French word . 
only one „ „ . 

two French words 

three „ „ 

four „ „ 

five2 „ „ 

Lines in the Prologue . 858 100-0 

If the total number of French words in the prologue be reckoned 
from the above data, they will be found to be 761, or not quite one 
word in a line on an average. The overpoweringly English character 
of the work could not be more clearly demonstrated. 

Chaucer's language may then be described as a degraded Anglo- 
Saxon, into which French words had been interwoven, without 
interfering with such grammatical forms as had been left, to the 
extent of about 20 per cent., and containing occasionally complete 
French phrases, of which, however, none occur in the prologue. 
To understand the formation of such a dead dialect, we have only 
to watch the formation of a similarly- constructed living dialect. 
Such a one really exists, although it must rapidly die out, as there 
are not only not the same causes at work which made the language 
of Chaucer develop into the language of England, but there are 
other and directly contrary influences which must rapidly lead to 
the extinction of its modern analogue. 

1 These are very few in number, see Mawr' or of Saynt Beneyt. 173, in 
5, 162, 254, 336, 429, 430, 646, 662. which the French words were in- 

2 The line is: The reuV of Saynt dispensable. 

. 325, 

per cent. 


. 343, 


. 157, 


. 87, 


. 12, 


. 1, 




Chap. VII. § 1. 

Pennsylvania German the Analogue of Chaucer's English. 

Fully one half of the people of Pennsylvania and Ohio in the 
United States of America understand the dialect known as Penn- 
sylvania German. This neij>hbourhood was the seat of a great Ger- 
man immigration from the Palatinate of the Khine' and Switzer- 
land. Here they kept up their language, and established schools, 
which are now almost entirely extinct. Surrounded by English of 
the XVII th century they naturally grafted some of its words on 
their own, either as distinct phrases, or as the roots of inflections ; 
and, perhaps, in more recent times, when fully nine-tenths of the 
present generation are educated in English, the amount of intro- 
duced English has increased.'^ The result is a living dialect which 
may be described as a degraded^ High German, into which English 

^ See supra, p. 47, lines 5 to 15. 

2 Some of these particulars have 
been taken from the preface to Mr. E. 
H. Ranch's Pennsylvanish Deitsch ! 
De Breefa fum Pit Schwefflebrenner un 
de Bevvy, si Fraw, fun Schliffletown 
on der Drucker fum " Father Abra- 
ham," Lancaster, Pa., 1868, and others 
from information kindly furnished me 
by Rev. Dr. Mombert, Lancaster, Penn- 
sylvania, U.S., in April, 1869. 

3 This does not mean that it is a 
degraded form of the present literary 
high German, but merely of the high 
German group of Germanic dialects. 
On 19 Aug. 1869, the 14th meeting of 
the German Press Union, of Pennsyl- 
vania, U.S., was held at Bethlehem, 
when an interesting discussion took 
place on Pennsylvania German, or das 
Deutsch-Pennsylvanische, as it is termed 
in the Reading Adler of 31 Aug. 1869, 
a German newspaper published at 
Reading, Berks County, Pa., U.S., fi'om 
which the following account is trans- 
lated and condensed. Prof. Notz, of 
Allentown, who is preparing a Penn- 
sylvania German grammar, drew at- 
tention to the recent German publi- 
cations on Prankish, Upper- Bavarian, 
Palatine, Swabian, and Swiss dialects, 
and asserted that the Penn. Germ, had 
an equally tough existence {zdhes Leben) 
and deserved as much study. Mr. Dan 
E. Schodler declared that the Germans 
of Pennsylvania could only be taught 
literary high German, in which their 
divine service had always been con- 
ducted, by means of their own dialect. 
Dr. G. Kellner justified dialects. He 
considered that linguists, including J. 
Grimm, had not sufficiently compre- 
hended the importance of dialects. 
Speech was as natural to man as walk- 

ing, eating, and drinking, and the 
original language of a people was dia- 
lectic, not literary, which last only 
finally prevailed, to use Max M tiller's 
expression as the high language, {Hoeh- 
sprache). The roots of a literary 
language were planted in its dialects, 
whence it drew its strength and wealth, 
and which it in turn modified, polished 
and ennobled. Was Penn. Germ, such a 
dialect ? Many English speakers, who 
knew nothing of German dialects, 
might deny it, and so might even many 
educated north Germans, who were un- 
acquainted with the south German 
dialects, and regarded all the genuine 
southern forms of Penn. Germ, as a 
corrupted high German, or as idioms 
borrowed from the English. They 
would therefore style it a jargon, not a 
dialect. Certainly, the incorporation 
of English words and phrases had given 
it some such appearance, but on re- 
moving these foreign elements it re- 
mained as good a dialect as the Alsa- 
tian after being stripped of its Gal- 
licisms, in which dialect beautiful 
poems and tales had been written, 
taking an honourable position in Ger- 
man literature. Penn. Germ., apart 
from its English additions, was a south 
German dialect, composed of Prankish, 
Swabian. Palatine, and Alle manic, 
which was interlarded with more or 
less English, according to the counties 
in which the settlements had occurred ; 
in some places English was entirely 
absent. AH that marked a dialect in 
Germany was present in Penn Germ., 
and since new immigration was per- 
petually introducing fresh high Ger- 
man, the task would be to purify the 
old dialect of its English jargon, and use 
the result for the benefit of the people 

Chap. VII. § 1. 



words have been interwoven, without interfering with such gram- 
matical forms as had been left, and containing occasionally complete 
English phrases. On referring to the first sentence of the last 
paragraph, the exact analogy of Pennsylvania Dutch to Chaucer's 
English will be at once apprehended. The dialect is said to possess 
a somewhat copious literature, and it is certainly an interesting 
study, which well deserves to be philologically conducted.^ Eor 
the present work it has an additional special value, as it continually 
exhibits varieties of sound as compared with the received high 
German, which are identical with those which we have been led to 
suppose actually took place in the development of received English, 
as (oo, ee, aa) for {aa^ ai, au). 

The orthographical systems pursued in writing it have been two, 
and might obviously have been three or more. The first and most 
natural was to adopt such a German orthography as is usually 
employed for the representation of German dialects, and to spell 
the introduced English words chiefly after a German fashion. This 
is the plan pursued, but not quite consistently,^ in the following 
extract, for which I am indebted to Dr. Mombert. The English 
constituents are italicised as the French are in the following edition of 
the prologue. A few words are explained in brackets [], but any one 
familiar with German will understand the original, which seems to 
have been written by an educated German familiar with good English. 

of Pennsylvania. The Penn. Germ, 
press was the champion of this move- 
ment, by which an entire German 
family would he more and more im- 
bued with modern German culture. 
As a striking proof of the identity of 
Palatine with Pennsylvanian German, 
he referred to Nadler's poems called 
Frohlich Pfalz, Gott erhaW s, which, 
written in the Palatine dialect, were, 
when read out to the meeting by Dr. 
Leisenring, a born Penn. German, as 
readily intelligible to the audience as if 
they had been written in Penn. German. 
Prof. Notz also observed that in Ger- 
many the people stni spoke among one 
another in dialects, and only excep- 
tionally in high German when they 
spoke with those who had received a 
superior education — and that even the 
latter were wont to speak with the 
people in their own dialect. This was 
corroborated by Messi's. Eosenthal, 
Hesse, and others. On the motion of 
Prof. Notz, it was resolved to prosecute 
an inquiry into the Germanic forms of 
expression in use in Pennsylvania, and 
to report thereon, in order to obtain 
materials for a complete characterisa- 
tion of the dialect. 

^ Prof. S. S. Haldeman, of Columbia, 
Pennsylvania, to whom I have been 

under great phonetic obligations, and 
who has been familiar with the dialect 
from childhood, has promised to fur- 
nish the Philological Society with 
some systematic account of this pecu- 
liar hybrid language, the living repre- 
sentation not only of the marriage of 
English with Norman, but of the 
breaking up of Latin into the Romance 
dialects. The Rev. Dr. Mombert, for- 
merly of Lancaster, Pennsylvania, but 
now of Dresden, Saxony, who has long 
been engaged in collecting specimens, 
has also promised to furnish some ad- 
ditions. The precedmg note shews the 
interest which it is now exciting in 
its native country. In this place it is 
only used as a passing illustration, but 
through the kindness of these com- 
petent guides, I am enabled to give 
the reader a trustworthy account so 
far as it goes. 

2 Thus vy is used for ee in keyn = 
(k^m), or rather (ketin) according to Dr. 
Mombert, and ee for ih (ii) in Teer, which 
are accommodations to English habits. 
Cow skin retains its English form. A 
more strictly German orthography is 
followed in L. A. Wollemveber' s Ge- 
malde aus dem Pennsylvanischen Volks- 
leben, Philadelphia und Leipzig, 1869, 
p. 76. 



Chap. YII. § 1. 

Ein Gespriich. 

1. Ah, Ddvee, was hot Dich 
gestern Owent [Abend] so ver- 
tollt schmdrt aus Squeier Esse- 
beises kumme mache ? War 
ebbes [etwas] letz ^ ? 

2. Nix apartiges ! ich hab 
jusht a bissel mit dor Pally 
gesp'drkt [played the spark], als 
Dir ganz uiivermuth der olte 
Mann derzu kummt, ummer 
[und mir] zu vershte* gibt, er 
dat des net ^leicher 

1. Awer [aber] wie hot er's 
dir zu vershteh' gegewe' (gege- 
ben] ? Grob oder hollich ? 

2. Ach net [nicht], er hat 
keyn [kein] wort geschwiitzt. 

1. Well, wie hot er's dann 
g'mocht ? 

2. Er hat jusht de Teer 

[Thiire] ufg'mocht, mir mei' 
Hutli iu de Hand 'gewe' un' de 
Coivskin von der Wand g'kricht 
[gekriegt]. Do hob' ich g'denkt, 
er that's net glei'che, dass ich die 
Pally shpdrke that un bin grod 
fortgange ; des wer alles, Sam. 

1 . Ja, geleddert hot er Dich, 
Davee, dann du bist net gauge, — 
g'shprunge bischt Du als wenn 
a dutzend Hund hinnig [hinter] 
Dich her wiiren. Ich hab dich 

wohl geseyhne [gesehen]. 

2. Well, sei nur shtill drfon 
[davon], und sags Niemand, 
sonst werd' ich ausgelacht. 

Sam versprach's ; awer som- 
how muss er sich doch ver- 
schnappt hawe [haben], sonst 
hatt's net g'druckt werde konne. 

The second style of orthography is to treat the whole as English 
and spell the German as well as the English words, after English 
analogies. This apparently hopeless task,^ was undertaken by Mr. 
E-auch, who in his weekly newspaper, Father Abraham, has weekly 
furnished a letter from an imaginary Pit i.e. Peter Schwefflebrenner, 
without any intei'pretation, and in a spelling " peculiarly his own."* 
Perhaps some of the popularity of these satirical letters is due, as 

^ South German letz, letsch, I'atsch, 
wrong, left-handed, as in high German 
links, for which Prof. Haldeman refers 
to Stalder, and to Ziemann, Mittel- 
hochdeutsches Worterb. 217. See also 
Schmeller, Bayerisches Worterb. 2, 
530, " (Miar is ktz) mir ist nicht recht, 
d. h. iibel." Compare high German 
vet^letzen, to injure. 

2 Dr. Mombert considers gleichen in 
this sense of "like, approve of," to be 
the EngHsh word like Germanized. 
But Dr. Stratmann, on seeing the 
passage, considered the word might be 
from the old high German lichen, to 
please. This verb, however, was in- 
transitive in all the Germanic dialects, 
and in old English (see Prol. 777 
below : if you liketh, where you is of 
course dative). The present active use 
seems to be modern English, and I 
have therefore marked it accordingly. 

3 An attempt of Chaucer's scribes to 
write his language after Norman ana- 
logies, as Rapp supposes to have been 
the case, would have been precisely 
analogous. Fortunately this was not 
possible, supra p. 588, n. 4, or we 
might have never been able to recover 
his pronunciation. 

* In the prospectus of his newspaper, 
Mr. Rauch says : "So weit das mer 
wissa, is der Pit Schwefflebrenner der 
eantsich moun in der United States 
d.iers Peimsylvanish Dcitsh recht shreibt 
un bushtaweert exactly we's g'shwetzt 
im ous g'shprocha wiierd," i.e.^ as far 
as we know. Pit Schwefflebrenner is 
the only man in the United States 
who writes and spells Pennsylvania 
German correctly, exactly as it is gos- 
sipped and pronounced. 


Chap. VII. § 1. 



some of the fun of Hans Breitmann's Ballads^ certainly is, to the 
drollness of the orthography, which however furnishes endless diffi- 
culties to one who has not a previous knowledge of the dialect.^ 

The third orthography would be the usual high German and 

1 Hans Breitmann's "poems are writ- 
ten in the droll broken English (not to 
be confounded with the Pennsylvanian 
German) spoken by millions — mostly 
uneducated — Germans in America, im- 
migrants to a great extent from south- 
ern Germany. Their English has not 
yet become a district dialect ; and it 
would even be difficult to fix at present 
the varieties in which it occurs." — 
Preface to the 8th edition of Hans 
Breitmann's Party, with other Ballads, 
by Charles G. Leland, London, 1869, 
p. xiii. In fact Mr. Leland has played 
with his dialect, and in its unfixed con- 
dition has made the greatest possible 
fun out of the confusion of p with h, t 
with d, and g with k^ without stopping 
to consider whether he was giving an 
organically correct representation of 
any one German's pronunciation. He 
has consequently often written combi- 
nations which no German would na- 
turally say, and which few could, even 
after many trials, succeed in pronoun- 
cing, and some which are scarcely 
attackable by any organs of speech. 
The book has, therefore, plenty of vis 
comica, but no linguistic value. 

2 The following inconsistencies 
pointed out by Prof. S. S. Haldeman, 
are worth notice, because similar ab- 
surdities constantly occur in attempts 
to reduce our English dialects, or 
barbaric utterances, to English analo- 
gies, by persons who have not fixed 
upon any phonetic orthography, such 
as the Glossotype of Chap. VI., § 3, 
and imagine that the kaleidoscopic 
character of our own orthography is 
not a mere " shewing the eyes and 
grieving the heart." Prof. H. says : 
" The orthography is bad and incon- 
sistent, sometimes English and some- 
times German, so that it requires some 
knowledge of the dialect, and of English 
spelling to be able to read it. 

" The vowel of they occurs in ferstay, 
m«A, nay, ehns, hoes and base ( = base, 
angry), li<«st {=heisst, called) mwich, 
daet, gea — m being mostly used (as in 
heasa, tswea) ; but gedreat (also dreet) 
rhymes its English form treaty and 
dreat, {=dreht, turns) with fate. 

" The German a is as in what and 
f«ll, but the former falls into the vowel 
of hwt, hut. Fr/11 is represented by ah 
in betz^Ala, and aa in p.'?ar, but usually 
by aw [au in saugd) as in aw {anch, 
also) g'sawt (said, gesagt). IIaiva = 
haben, should have been haw-wa. The 
vowel of what is represented by a or 
0, as in w«s, wot, hflib, k«nn, donn, 
norra, gonga, 

" of no occurs in boAna, so amoAl, 
= einmal, co«xa (=to coax!) doch, 
ho«r { = haar hair), woch, iroke. 

" When German a has become Eng- 
lish II of hut, it is written u, as in hict 
( = hat, has) , and a final, as in macha, 
denka = denken, [which = (-e)], an = ein. 

" The vowel of f/eld occurs in w«V, 
shp2Vla, dip, sh-'^-s, kr^ya = (krii<7hT;), y 
is used throughout for {g]i) of regen. 
The y of m// occurs in sei, si, my and 
mei, hei, dyfel, subscr/ba. 

" TF, when not used as a vowel, has 
its true German power (bh), as in 
tstvea = ztvei, h.&.ic'ii = habe>t, i6'easht = 
weisst, ^^enich and weanich ! = wenig^ 
SLwer = aber, and some other examples 
of b have this sound. 

" Das is for dass that, and des is 
used for the neuter article das. The s 
is hissing (s). The r is trilled (.r) as 
in German. P b, t d, k g, are con- 
fused. The lost final n is commonly 
recalled by a nasalised vowel. 

" Oo in fool, full, appears in un, 
when used for lotd, uf for anf, wu = 
vjo where, Zcitting pure German, shoola 
= schools. fyuvel = iYo\\h\Q. 

" English words mostly remain Eng- 
lish in pronunciation, as in : meeting- 
house, town, frolic, for instance, horse- 
race, game poker shpeela, bensa pitcha 
= pitch pence, uf course; but many 
words are modified when they cross a 
German characteristic, thus greenbacks, 
the national currency, is rather (kriin*- 

"The vowel of fat occurs in 
Barricks = Berks county, lodwarrick 
lodwserrick = lativerge electuary, kger- 
rich = kirche, wfert = iverth, h«r = her. 
-le is only an English orthography for 
el or 'I, sh is English." 



Chap. VII. § I. 

English orthographies for the words used, which would of course 
convey no infunnation respecting the real state of the dialect. The 
only proper ortliography, the only one from which such information 
can be derived, is of course phonetic. The kindness of Prof. Halde- 
mann has enabled me to supply this gieat desideratum.* The 
passage selected is really a pulf of a jeweller's shop in Lancaster, 
Pa., and was chosen because it is short, complete, characteristic, 
varied, and, being not political, generally intelligible. It is given 
first in ^Ir. llauch's peculiar Anglo-Gennan spelling, and then in 
Prof. Haldemann's phonetic transcript, afterwards by way of ex- 
plaining the words, the passage is written out in ordinary High 
German and English, the English words being italicised, and finally 
a verbal English translation is furnished. On pp. 661-3 is added 
a series of notes on the peculiarities of the original, referred to in 
the first text. The reader will thus be able to form a good idea of 
the dialect, and those who are acquainted with German and English 
will thoroughly appreciate the formation of Chaucer's language. 

^ Professor Haldeman not having 
spoken the dialect naturally for many 
years, after completing his phonetic 
transcript, saw Mr. Ranch the author, 
and ascertained that their pronuncia- 
tions practically agreed. The phonetic 
transcript, here furnished, may there- 
fore be relied on. Prof. Haldeman 
being an accomplished phonetician, and 
acquainted with my palaeotype. wrote 
the pronunciation himself in the letters 
here used. Of course for publication 
in a newspaper, my palaeotype would 
not answer, but my glossotype would 
enable the author to give his Penn- 
sylvania German in an English form 
and much more intelligibly. Thus the 
last paragraph in the example, p. 661, 
would run as follows in glossotype, 
adopting Prof. Haldeman' s pronuncia- 
tion : '* Auver iyh kon der net ollas 
saugha. Va'rr [vehrr] mai'uer vissa 
vil, oonn va-rr [vehrr] farrst raiti 
Krishtaukh sokh vil— dee faaynsti oonn 
beshti bressents, maukh selverr dorrt 
ons Tsaums gaia, oonn siyh selverr 
soota. Noh mohrr et press'nt. Peet 
Shveif'lbrennerr." But the proper 
orthography would be a glossotype 
upon a German instead of an English 
basis. The following scheme would 
most probably answer all purposes. 
The meaning of the symbols is ex- 
plained by German examples, unless 
otherwise marked, and in palaeotype. 
Long vowels : ie btb (ii), ee heei {ee), 
ae sprache (ee, seae), aa Aa\ (aa), ao 
Eng, awl (aa), oo Boot (oo), uh Pfj^AI 

(uu), ue UehQl (yy), oe Oe\ (ceoc). 
Short Vowels : i iSmn (i, «i), e B<;tt 
(e, e), a Eng, b«t (e, ae), a «11 {a), a 
Eng. what (a o), o Motte (o o), w Pfwnd 
(u, ^<), u Fwlle (y), o Bocke (oe), e eine 
(b), Eng. hui (b, a), (J sign of nasality. 
Diphthongs : ai Ham (ai), oi Eng. 
joy, Hamburgh Eu\q (oi), aii theo- 
retical Eu\q (ay), au kawen (an). 
Consonants : j Ja (j), w wie i^bh), 
Eng. w (w) must be indicated by a 
change of type, roman to italic, or con- 
versely, k heu (u)^ p b (p b), ^ <^ (t d), 
tsch dsh (tsh dzh), k ff (k. g), ku (kn), 
/ V (f v), th dh (th dh), ss Niisse (s), 
s wiese (z), sch sh (sh zh), ch gh (Jcii 
kh, ^h gh), rim n (rim n), ng nk 
(q qk). German readers would not 
require to make the distinction ss, s, 
except between two vowels, as Wiese, 
Niisse, Fuesse. They would also not 
find it necessary to distinguish between 
e, e final, or between er\ er, unaccented. 
For similar reasons the short vowel 
signs are allowed a double sense. This 
style of writing would suit most dia- 
lectic German, but if any additional 
vowels are required ih, eh, ah, oh, are 
available. The last sentence of the 
following example, omitting the dis- 
tinction e, e, would then run as fol- 
lows : " Aower ich kon der net olles 
saoghe. Waer meener wisse wil, un 
waer ferst recti Krischtaoch sokh wil, 
— die fainsti un beschti bressents, maokh 
selwer dort ons Tsaoms geee, un sikh 
selwer suhte. Noo moor et press'nt. 
Piet Schwefflbrenner." 

Chap. VII. § 1. 




Kauch's Orthography. 
Pennsylvanish Deitsh. 

Mr.^ Podder Abraham^ Printer 
— Deer Sir : Ich kon mer now 
net^ helfa* — ich mus der yetz 
amohP shreiva^ we ich un de 
Bevvy^ ousgemocht hen doh fer- 
gonga^ we mer in der shtadt 
Lancaster wara. 

Der hawpt^ platz wu^° mer 
onna^^ sin, war dort in selly 
Zahm's ivver ous sheana Watcha^^ 
un Jewelry establishment, grawd 
dort om eck^^ fun was se de Nord 
Queen Strose^* heasa un Center 
Shquare — net weit fun wu das 
eier office is. 

In all meim leawa hab ich ne 
net so feel tip-top sheany sacha 
g'sea, un selP^ is exactly was de 
Bevvy sawgt.^^ 

We mer nei sin un amohl so a 
wennich rum geguckt hen, donn 
secht^^ de Bevvy — loud genunk" 
das der monn 's hut heara kenna 
— " Kow Pit," '8 secht se, "weil 

Prof. Haldeman's Pronunciation. 
Peus^'Iv^^ •n«sh D^^itsh. 

M^s•t'r FAd''r lAA-brahAm 
prm't'r — Diir Sor : Ikh. kAn m'r 
n^u net helf'^ — ikh. mus d'r jets 
VTROol' shraibh-B bhii ^7•h un di 
Bebh'i ^us'gcmAkht Hen doo 
f'rgAq'i] bhii m'r m d'r shtAt 
Leq'kesht'r bhAATB. 

D'r HAApt plAts bhuu m'r au'b 
sm, bliAr dArt m sel•^ TsAAms 
ibh''r aus shee'iiTe bhAtsh"B un 
tshu-Blr^' estep'l/shmBut, grAAd 
dArt Am ek fun bhAs si di ^ort 
Kfiin Shtroos uee's^ un Sen-t'r 
Shkbh^^r — net wait fun bhuu 
dAs ai*'r Af*/s is. 

In a1 jRaim. leehh-yi HAb ikh. 
nii net so fiil t^p'tAp shee'ni 
sAkh'B ks^(9*^ un sel is eksaek'k' 
bhAs di Pebh-e sAAkt. 

Bhi m'r nai sm un •r?mool soo 
T3 bhen'«X'h rum gegukt' Hen, 
dAn seAht di Bebh"? — kut gn- 
nuqk" dAS d'r mAns not neer'B 
ken-B — *'K«u P^'t," se^ht si. 

3. German and English Translation, 

4. Verbal English Translation, 

Pennsylvanisches Deutsch. Pennsylvania German. 

Mr. Vater Abraham, Printer — Bear 
Sir : Ich kann mir now nicht helfen — 
ich muss dir jetzt einmal schreiben wie 
ich und die Barbara ausgemacht haben, 
da vergangen, wie wir in der Stadt 
Lancaster waren. 

Der Haupt-Platz wo wir an sind, 
war dort in selbiges Zahms iiberaus 
schone JFatche und Jewelry Estab- 
lishment, grade dort an-der Ecke von 
was sie die Nord Queen Strasse heis 
sen und Centre Square — nicht weit von 
wo dass euer office ist. 

In all meinem Leben habe ich nie 
nicht so viele tiptop schone Sachen 
gesehen, und selbiges ist exactly was 
die Barbara sagt. 

Wie wir hinein sind und einmal so 
ein wenig herum geguckt haben, dann 
sagte die Bai'bara — laut genug dass der 
Mann es hat horen konnen — "iVow;, 

Mr. Father Abraham, Printer — 
Bear Sir : I can myself now not help 
— I must to-thee now once write, how I 
and the Barbara managed [i.e. fared] 
have there past, as we in the town 
Lancaster were. 

The chief-place where we arrived 
are, was there in same Zahm's over- 
out beautiful Watches and Jewelry 
Establishment, exactly there at corner 
of what they the North Queen Street 
call, and Centre Square — not far from 
where that your office is. 

In all my life have I never not so 
many tiptop beautiful things seen, and 
same is exactly what the Barbara 

As we hence-into are, and once so a 
little around looked have, then said the 
Barbara — loud enough that the man it 
has to-hear been-able — "Now, Peter" 



Chap. VII. $ 1. 

1. Ranch* 8 Orthography, continued. 
se der di watch g'shtola hen 
dort in Nei Yorrick,^^ musht an 
neie kawfa, un doh gookts das^^ 
wann^" du dich suta'^' kennsht."'^^ 

We se sell g'sawt hut, donn 
hen awer amohl de karris '^^ dort 
hinnich^* cm counter uf geguckt. 
Eaner hut si brill gedropt,^^ 
un an onnerer is uf g'shtonna 
un all hen mich orrig^^ freind- 
lich aw^^ geguckt. 

Donn sogt eaner — so a wen- 
nich an goot guckicher^® ding — 
secht er, '' Ich giawb doch now 
das ich weas waer du bisht." 
** Well, " sog ich, '' waer 
denksht ?" '' Ei der Pit Schwef- 
flebrenner." '' Exactly so," hab 
ich g'sawt. '' Un des doh is 
de Bevvy, di alty," secht er. 
** Aw so," hab ich g'sawt. 

Donn hut er mer de bond 
gew^a, un der Bevvy aw, un 
hut g'sawt er bet shun feel fun 
meina breefa g'leasa, un er waer 
orrig froh mich amohl selwer 

3. Germ. ^ Eng. Translation, cont. 

Peter,*^ sagte sie, "weil sie dir deine 
Watch gestohlen haben dort in Neu 
York, musst du eine neue kaufen, and 
da guckt es [als] dass wann du dich 
stiiten konnest." 

"Wie sie selbiges gesagt hat, dann 
haben aber einmal die Kerls dort hin- 
terig dem counter aufgeguckt. Einer 
hat seine Brille gedropt, und ein an- 
derer ist aufgestanden und alle haben 
mich arg freundlich angeguckt. 

Dann sagt einer — so ein wenig ein 
gutguckiges Ding — sagte er, " Ich 
glaube doch now dass ich weiss wer du 
hist." ''Well," sage ich, "wer 
denkest?" "Ei, der Peter Schwefel- 
brenner." " Exactly so," habe ich 
gesagt. " Und das da ist die Barbara, 
deine Alte," sagte er. " Auch so," 
habe ich gesagt. 

Dann h t er mir die Hand gegeben, 
und der Barbara auch, und hat gesagt 
er hatte schon viel von meinen Briefen 
gelesen, und er ware arg froh mich 

2. Haldeman's Pronunciation , cont. 
**bhail si dir d^i^ bhAtsh 
kshtool'i} nen dArt m N«i JAr-^k, 
musht Tjn nai'M kAAf'c, un doo 
gukts dAS bhAn du Oiikh. suut"^ 

Bhi si sel ksAAt not, dAn Hen 
AA'b'r "Bmool' di kaerls dArt Hm*- 
ikh. i?m kaunt"'r uf g^gukt*. 
^6;'n'r not ^ai bril gcdrApt', un 
en AH'^rar «s uf kshtAn'B un a1 
Hen Tciikh. Ar'^'kh fr^ind'h'Ah aa^ 

Dah sAkt (?(?'n'r — soo 'B bhen•^/;h 
-en guut guk'/A:h'r d«'q — se^ht ■er, 
''/A-h glAAb doA'h n«u dAs ikh 
bhe^s bhaer du b/sht." *' Bhel," 
sAg ik\ "bhaer deqksht?" " Ai 
d'r Pit Shbh^efibren-'r." " Ek- 
saek'k' soo, ' ' HAb ikh. ksAAt. " " Un 
des doo «s di Bebh**, dai Alt**,'' 
seX'ht aer. *' :Aa soo," HAb ikla. 

Dau Hat aer m'r di HAnd 
gebh'B, un d'r Pebh'* aa, un nat 
ksAAt aer net shun fill fun mam"^ 
briif'a ^ee'm, un aer bhaeaer 
M'ikh. froo inikh. -Bmool* sel'bh^r 

4. Verbal Eng. Translation, cont. 

said she, "because they to-thee thy 
watch stolen have there in New York, 
must thou a new (one) buy, and there 
looks it [as] that if thou thee suit 

As she same said has, then have 
again once the fellows there behind the 
counter up-looked. One has his spec- 
tacles dropped, and another is up -stood, 
and all have me horrid friendlily on- 

Then says one — so a little a good- 
looking thing — said he, "I believe, 
however, now that I know who thou 
art." " Well," say I, " who thinkest 
(thou that I am) ? " " Eh, the Peter 
Sulphurburner." " Exactly so," have 
I said. " And that there ist the 
Barbara, thy old-woman," said he. 
" Also so," have I said. 

Then has he me the hand given, and 
to-the Barbara also, and has said he 
had already much of my letters read, 
and he was horrid glad me once self to 

Chap. VII. § 1. 



1. RaucKs Orthography, continued. 
tsu seana.^^ Dorm sin mer awer 
amolil on bisness. 

Watcha hen se dort, first-raty 
for 16 dahler bis tsu 450 dahler. 
ISToch. dem das mer se amohl 
recht beguckt hen, is de Bevvy 
tsu der conclusion kumma an 
Amerikanishe watch tsu kawfa. 

Dort hen se aw was se Ter- 
mommiters heasa — so a ding 
dass eani^° weist we kalt s' wetter 
is, un sell dinkt mich kent mer 
braucha alleweil. Any-how mer 
hen cans gekawft. 

De watch is aw an first-raty. 
Ich war als^^ uf^^ der meanung 
das de Amerikanishe watcha 
waerra drous in Deitshlond 
g'macht, un awer sell is net 
wohr. Un de house-uhra ; chee- 
many^^ fires awer se hen about 
sheany ! TJf course mer hen aw 
eany gekawft, for wann ich 
amohl Posht Meashder bin mus 
ich eany hawa for^^ in de office 
ni du. 

3. Germ. ^ Eng. Translation, cont. 

einmal selber zu sehen(en). Dann sind 
wir aber einmal an business. 

Watehe haben sie dort, first-rate-^ 
fiir sechzehn bis zu vier hundert (und) 
fUnfzig Thaler. Nachdem dass wir sie 
einmal recht beguckt haben, ist die 
Barbara zu der conclusion gekoramen 
sine Amerikanische watch zu kaufen. 

Dort haben sie auch was sie Ther- 
mometers heissen — so ein Ding das 
einem weiset wie kalt das Wetter ist, 
und selbiges diinkt mich konnten wir 
brauchen alleweile. Anyhow wir 
haben eines gekauft. 

Die Watch ist auch eine first-rate-e. 
Ich war also auf [alles auf, also of?'] 
der Meinung dass die Amerikanischen 
Watehe waren draussen in Deutschland 
gemacht, und aber selbiges ist nicht 
wahr. Und die Hausuhren ; Gemini 
fires ! aber sie haben about schone ! Of 
course wir haben auch eine gekauft, 
for wann ich einmal Post blaster bin, 
muss ich eine haben for in die office 
hinein [zu] thun. 

2. Haldeman's Pronunciation, cont. 
tsu B>een."^. Dau sm m'r AAbh''r 
"Bmool* An b/s'U'Bs. 

BhAtsh"B Hen si dArt, farst 
mee'ti f'r se/jh'tSf?^ bzs tsu fiiv- 
Hun'^rt-fuf'tszX'h tAAl'^r. JS^Akh 
dem dAs m'r sii "evaooV reZ;ht 
bT3gukt* Hen, «s di Pebh•^ tsu d'r 
kAnklun'sh^n kuni'i? ^n :Amere- 
kAA"n/shii bhAtsh tsu kAAf*^. 

DArt Hen si aa bhAs si ter- 
mAm'^'t'rs h^^s'a — so 'b d«qdAs eevci 
bhaist bhi kAlt 's bhet*'r /s, un 
sel d?qt m^/(•h kent m'r br«ukh"B 
Al'abhrtil. En'mau m'r Hen 
een^ gi^kAAft*. 

Dii bhAtsh z's aa ^n farst ree'ii. 
//h bhAr aIs uf der m6<?'nuq dAs 
dii :Amer/kAA'nisht? bhAtsh "b 
bhaer*^ (\ia\xs, m D^itsh'lAut 
gmAAkht*, un AA'bh'r sel «s 
net bhoor. Un dii h^us'uutb ; 
tshii•ml]n^ f^irs ! AA'bh'r si Hen 
•eb^ut* sh^^'ne! Uf koors m'r 
Hen AA een'i gckAAft', f'r bhAn 
ik\i 'Bmool* Poosht Meesh't'r b/n 
mus ikh. ee'm HAA'bh^ for m di 
Af•^s na\ du. 

4. Verbal Eng. Translation, cont. 

see. Then are we again once on 

Watcher, have they there, first-rate 
(ones) for sixteen up -to four hunderd 
(and) fifty dollars. After that wie 
them once rightly beseen have, is the 
Barbara to the conclusion come, an 
American watch to buy. 

There have they also what they 
Thermometers call — so a thing that 
to-him shows how cold the weather 
is, and same thinks me might we use 
presently. Anyhotv we have one 

The watch is also a first-rate (one). 
I was always on [all up = entirely 
of, always of] the opinion that the 
American tvatches were there- out in 
Germany made, and but same is not 
true. And the houseclocks ; Gemini 
Fires ! but they have aboict beautiful 
(ones) ! Of course we have also one 
bought, for when I once Post Master 
am, must I one have, for into the 
office hence-in (to) do. 



Chap. VII. § 1, 

1. Mauch's Orthography y continued. 

Se hen aw an grosser shtock 
fun Silverny Leffla, Brilla, un 
ich weas net was olles. Do 
Bevvy hut geclu das well ich 
yctz boll amoliP'* an United 
Shtates Government Officer si 
waer, set ich mer aw an Brill 
kawfa, un ich hab aw eany 
krickt das ich now net gewa 
deat fer duppelt's geld das se 
gekosht hut, for ich kon yetz 
noch amohl so goot seana un 
leasa das^^ tsufore. 

Un we ich amohl dorrich my 
neie Brill geguckt hab, donn 
hab ich aersht all de feiny sacha 
recht beguckt, un an examina- 
tion gemacht fun Breast Pins, 
Rings, Watch-ketta,^'^ Shtuds, 
Messera un Gowella, etc. 

Eans fun sella Breastpins hut 
der Bevvy about goot aw-g'- 
shtonna, awer er hut mer doch a 
wennich tsu feel g'fuddert der- 
fore — 25 dahler, un donn hab 

2. Haldeman'a Pronunciation^ cont. 

Sii Hen A.i isn groo'SB shtxk 
fun SA-bliBrn? Lef-'le, Br/l'B un 
ilch. hhee^ net bhAs aI'bs. Dii 
Pebh'« Hat gnduu* dAs bh^il i^h 
Jets bAl 'Gmool' T?n Junai't^t 
Shte(9ts Gofrmunt Of*/ser mi 
bhucaer, set ikh. m'r a a un Bnl 
kAA'fB, un ikh HAp aa ee'ni krekt, 
dAS ikh. n«u net gebh'is deet f r 
dup-'lts geld dAS sii gekosht* 
net, f'r ikh. kAn jets nokh Timool' 
soo guut me'jre un lee'^^e dAs 

Un bhii ilch. 'emool' dAT'e^h 
mrtij nai'i Br«l gegiikt* HAp, 
dAn HAp ilch. aersht a1 dii fai'm 
sAkh*^ re^ht begukt* un v,n 
eks9emm(?sh-'n gumAkht* fun 
Bresht'pms, R«qs, BhAtsh'ket-'B, 
Shtots, Mes'BFB un GAbh-'lij, 

Eens> fun sel*e Bresht'pms not 
d'r Bebh** ■eb^ut* guut aa/- 
gsht-AAn"B, AA'bh'r cer not m^r 
dokh u bhen^X'h tsu fiil gfud''rt 
d'rfoor* — fmf un tsbhAH's^'kh 

3. Germ. ^ Eng. Translation, cont. 

Sie haben auch einen grossen stock 
von sUbernen LofFeln, Brillen, und ich 
weiss nicht was alles. Die Barbara 
hat gethan dass weil ich jetzt bald 
einmal ein United States Government 
Officer sein werde, sollte ich mir auch 
eine Brille kaufen, und ich habe auch 
eine gekriegt, dass ich now nicht geben 
thate fiir doppelt-das Geld das sie 
gekostet hat, for ich kann jetzt noch 
einmal so gut sehen und lesen [als] 
dass zuvor. 

Und wie ich einmal durch meine 
neue Brille geguckt habe, dann habe 
ich erst alle die feinen Sachen recht 
beguckt und an examination gemacht 
von Breastpins, Rings, Watch-\QiiQVi, 
Studs, Messer und Gabeln, etc. 

Eins von selbigen Breastpins hat der 
Barbara about gut angestanden, aber er 
hat mir doch ein wenig zu viel gefodert 
dafiir — fiinf und zwanzig Thaler — und 

4. Verbal Eng. Translation^ cont. 

They have also a great stock of silver 
spoons, spectacles, and I know not 
what all. The Barbara has done [es- 
timated] that because I now soon once 
a United States Government Officer be 
shall, should I me also a pair-of-spec- 
tacles buy, and I have also one got, 
that I now not give would-do for 
double the money that it cost has, for 
I can now still once so good see and 
read [as] that before. 

And as I once through my new 
spectacles looked have, then have I 
first all the fine things right be-seen, 
and an examination made of Breast- 
pins, Rings, Watchohdivas,, Studs, knives 
and forks, etc. 

One of the same Breastpins has the 
Barbara about good on-stood [suited], 
but he has me, however, a little too 
much asked therefore — five-and-twenty 

Chap. VII. § 1. 



1. Ranch's Orthography ^ continued. 

ich mer tsuletsht eany rous ge- 
pickt fer drei faertle dahler, fer 
seUy sogt de Bevvy, is anyhow 
ahead fun ennicher^^ onnery in 

Awer ich konn der net alles 
sawya. Waer meaner^^ wissa 
will, un waer first raty krishdog 
sach will — de feinsty un beshty 
presents, mog selwer dort ons 
Zahms gea un sich selwer suta. 
Ko more at present. 

Pit Schwefflebrenner. 

2. Haldeman's Prommciation, cont. 

tAA'l'r, un dAn HAb ilck ni«r 
tsuletsht' ee'ni r<?us gBpz.'kt* f r 
trai faer't'l tAA'l^r, f'r sel** sAkt 
di Bebh*/ «s en'm^u ■ehet' fun 
en"/>^h^r Kw^vi in Shl/f'lt«un. 

lAa'bb'r ikh kAn d'r net aI^bs 
SAA'gh^. Bhaer -meen'^m bh«s*B 
bh2l, un bhaer fi^rst Teei'i Kr/sh*- 
tAAkh sAkh bh/1 — dii ffl^in'sht* 
un besht"*' bres'ents, mAAkh sel'- 
bh'r dArt Ans Tsaauls ^ee"^ un 
^ikh. sel'bh'r suu'tu. Noo moor 
et bres''nt. 

Piit Shbhefibren-'r. 

3. Germ. ^ Eng. Translation, cont. 

dann habe ich mir zuletzt eine heraus 
gepickt fiir drei Viertel Thaler, for 
selbiges sagt die Barbara is anyhow 
ahead von einiger anderen in Schliffel- 

Aber ich kann dir nicht alles sagen. 
Wer mehr wissen will, und wer Jirst- 
rate-Q Christtag Sachen will — die 
feinsten und besten presents, mag selber 
dort an's Zahms gehen und sich selber 
suitQn. No more at present. 

Peter Schwefelbrenner. 

4. Verbal Eng. Translation, cont. 
dollars — and then have I for- me at- 
last one out picked for three-quarters 
(of a) dollar, for same says the Barbara 
is anyhow ahead of any other in 

But I can thee not all say. "Who 
more know will, and who first-rate 
Christmas things will — the finest and 
best presents., may himself there to-the 
Zahra's (house) go, and him self suit. 
No more at present. 

Peter Schwefelbrenner. 

Notes on the above Text. 

^ Mister is used as well as the 
German form (m^^sh-t'r). — S. S. 

2 Father Abraham means the late 
president Abraham Lincoln, assumed 
as the title of Ranch's newspaper. 

3 The guttural omitted, as frequently 
in nicht, nichts. 

* The infinitive -e for -en, as fre- 
quently in Chaucer, and commonly 
now on the Rhine. 

5 Einmal, a common expletive, in 
which the first syllable, even among 
more educated German speakers sinks 
into an indistinct («). Observe the 
transition of (a) into (oo). 

6 The common change of (b) into 

■^ Bevvy, or Pevvy, is a short form 
of Barbara, a rather common name in 
the dialect. Both forms are used in the 
following specimen. — S.S.H. German 
Babbe, Babchen, compare the English 
Bab, Babby. 

8 Doh here, fergonga recently, an 
advei'b, not for vergangene Woche. — 
S. S. H. 

^ Observe the frequent change of 
the German au, indisputably (au, <?u) 
into English (aa), precisely as we find 
to have occurred in English of the 
XVII th century. 

^° The not unfrequent changes of o 
long into (uu) are comparable to 
similar English changes xv th century. 

^^ Onna, the preposition an used as a 
verb, as in the English expression, 
" he ups and runs." I take this view 
because sind is an auxiliary and a 
present tense form, but the adverbial 
tendency of onyia (as if thither) must 
nevertheless not be overlooked. A 
German will sometimes use in English 
an expression like " outen the candle !" 
rarely heard in English — S.S.H. 

^2 Observe here a German plural 
termination e alfixed to an English 



Chap. VII. § 1. 

^3 Ecke being feminine, the correct 
form is an der Ecke^ although -tck in 
composition is neuter, as drtieck^ vier- 
eck. — S.S.II. In Schmeller's Bayr. 
Wort. 1, 25, " das Eck, eigentlich 
Egy" is recognized as south German. 
In the foUowing word fun for von^ 
short becomes (u) or («<). 

^* This change of German a to o is 
common, as in (shloof-u) for schlujhiy 
(shoo/) for schdf, etc. — S.S.H. See 
note 5, and compare this with the 
change of ags. {cut) into South English 
(00, 00), while (aa) remained in the 

^^ This frequent and difficult word 
has been translated selbiges throughout, 
as the nearest high German word, and 
selly, 9 lines above it, may, in fact, in- 
dicate this form . Compare Schmel- 
ler's Bayr. Wdrt. 3, 232, " Selb [de- 
clinabel] in Schwaben bfter nach erster 
Declin.-Art (seler, e, es), in A. B. 
lieber nach zweiter [der, die, das (s'l, 
den s'ln, di s'ln), etc.] gebraucht, statt 
des hochd. jener, e, es, welches un- 
volksliblich ist. [Fiir der, die, das 
selbe im hochd. Sinn. d.h. idem, eadem, 
idem, braucht die Mundart der die, 
das nemlicJie.] (s'l as m«l, dts s'l ma\, 
s'l'mfflz) jenes Mel, (s'l a tsait) zu 
jener zeit, (s'l at-Hwlb-m) oder (-bhfgq) 
des[jenigen] wegen." 

^6 Sawgt=sagt, says, secht = sdgt, 
instead of sagte, said, with the Umlaut. 
— S. S. H. The weak verb has there- 
fore a strong inflection. This distinc- 
tion is preserved throughout. Compare 
the common vulgar (and older ?) forms 
slep, sivcp, with the usual slept, wept, 
and see supra p. 355, art. 54. 

'' Genunk, with educed k, is com- 
mon in archaic and provincial German, 
and Rollenhagen rhymes jimg, pro- 
nouncedy?<;/cA: dialectically, with trunk. 
— S. S. H. See supra p. 192, n. 1. 

'^ (P/t) or (Piit) may be used for 
this short form of Peter. — S.S.H. It 
is the English Fete, not a German 
form as the voAvel shews 

^3 Observe the vowel educed by the 
strong trill of the ( r). For con- 
venience (r ) has been printed through- 
out, but the reader must remember 
that it is always distinctly, and some- 
times forcibly, trilled with the tip of 
the tongue, and never sinks to (.i). 

20 Las tvann, that though, as 
though. — S. S. H. Gookts das wann, 
for sieht es aus als oh, it looks as if. 
See note 36. 

21 Observe the German infinitive 
termination -e for -en, added to a 
purely English verb. 

2- The development of s into (sh) ig 
remarkable in high German. It is 
acknowledged as the proper pronun- 
ciation before t, p at the beginning of 
a syllable, throughout Germany, even 
North German actors not venturing to 
say (st-, sp-) even in Hamburg, as I 
am informed, the capital of that pro- 
nunciation. But in final -st, the 
common (-sht) is looked upon as a 
vulgarism, even in Saxony. 

23 Kcerls, may have an English «, 
but the form is often playfully used by 
good speakers in Germany, and hence 
may have been imported and not 

2* Hinnich for hinter has developed 
a final -ig, but this is a German ad- 

2' Gedropt, the German participial 
form for dropped. So also elsewhere I 
find gepimished, which may be com- 
pared with Chaucer's ypunish'd, Prol. 
V. 657. 

26 Orrig, very, Swiss arig (Stalder 
1, 110), German arg, but not used in 
a bad sense. — S.S.H. The word arg 
implies cunning and annoyance, but 
its use as an intensitive is comparable 
to our horrid, awfully, dreadfully, 
which are frequently used in a good 
sense, as : horrid beautiful, awfully 
nice, dreadfully crowded. Das ist zu 
arg ! that is too bad, too much ! is a 
common phrase even among educated 

2^ A IV for German an is nasalised, 
which distinguishes it from the same 
syllable when used for the German 
auch, also. — S. S. H. This recent 
evolution of a nasal sound in German, 
common also in Bavarian, may lead us 
to understand the comparatively recent 
nasal vowels in French, infra Chap. 
VIII, § 3. 

2*^ '1 he gender is changed because it 
refers to a man ; so in high German it 
is not unfrequent to find Frdulein, 
Miidehen, although they have a neuter 
adjective, referred to by a feminine 
pronoun, as : " das Fraulein hat ihren 
Handschuh fallen lassen," the young 
lady [neuter] has dropped her [fern.] 

29 In an earlier line g' sea iox gesehen, 
but here we have a double infinitive, 
as if zu sehenen. This is also used for 
the third person plui-al of the present 

Chap. VII. § 1. 



tense, as in sie gehen-a, they go. — 
S.S.H. Compare also ich hah dich, 
wohl geseyhne, in the Gesprdch, p. 
654. This seems comparable to what 
Prof. Child calls the protracted past 
participle in Chaucer, supra p. 357, 
art. 61. It is impossible to read 
the present specimen attentively with- 
out being struck by the similarity 
between this Pennsylvania German 
and Chaucer's English in the treat- 
ment of the final -e, -en of the older 
dialects. The form (sel-bher) in the 
preceding line preserves the b in the 
form (bh j . Schmeller also allows selber 
to preserve the b as (s'l'ba), see n. 15. 
3" Das earn tveist, that shews him, 
that shews to one or a person. — 
S. S. H. Eam.=einem, not ihm. 

31 This ah is Swiss, which Stalder 
defines by ehedem hitherto and immer 
always, compare ags. eal-enge altoge- 
ther and eal-wig always. — S.S.H. See 
also Schmeller Bayr.-Wort. 1, 50. Dr. 
Mombert takes als to be an obsolete 
high German contraction of alles in 
the sense of ever, mostly, usually. 

32 Prof. Haldeman takes uf for avf, 
but der Meir/ung, and not auf der 
Meinung, is the German phrase, and 
hence the word may be English, 
as afterwards, uf course. But this 
is hazardous, as uf in this sense could 
hardly be joined with a German dative 
der Meimmg. Can als uf be a dialec- 
tic expression for alles auf, literally all 
up, that is, entirely? Compare, Schmel- 
ler, Bayr. Wort. 1, 31, "«?(/ und auf, 
von unten (ganz, ohne Unterbrechung) 
bis oben, axif und nider vom Kopf bis 
zum Fuss, ganz und gar." 

^3 Cheemany is the English exclama- 
tion Oh jeemany. — S.S.H. The Eng- 
lish is apparently a corruption of: (Jh 
Jesus mihi, and has nothing to do with 
the Gemini. But what is the last part 
of this exclamation : Jires ? Prof. 
Haldeman, suggests, hell fires ! Dr. 
Mombert derives from the shout of: 
fire ! Can the near resemblance in 
sound between cheemany and chimney, 
have suggested the following fires ? 
Such things happen. 

31 For in de office ni du seems to 
stand for um in die office hinein zii 
thun. The use of for for um is a mere 
Anglicism, but why is zu omitted be- 
fore thun ? By a misprint, or dialec- 

tically for euphony ? It is required 
both by the German and English 
idiom. Dr. Mombert considers the 
omission of zu dialectic in this place, 
elswhere we find zu do. 

3= Boll amohl, bald einmal, pretty 
soon, shortly. This use of einmal once, 
appears in the English of Germans, as 
in : " Bring now here the pen once." 

36 Bas. This is not the neuter 
nominative article das, which is des in 
this dialect, but a contraction of als 
dass, with the most important part, 
als, omitted. — S.S.H. I am inclined 
to take it for dass used for als, as in 
the former phrase das wayin — als ob, 
see note 20. According to Schmeller, 
Bayr. Wort. 1, 400 " dass schliesst 
sich als allgeraeinste conjunction, in 
der Rede des Volkes, gern andern con- 
junctionen erklarend an, oder vertritt 
deren Stelle.'" 

^"^ Watch-ketta, a half English, half 
German compound, is comparable to 
Q,\\im(iQx'' ii foot mantel, half English and 
half French, in Prol. infra, v. 472, and 
supra p. 651, 1. 6. 

3^* This may be the English any. 
like the German einig, treated like 
einiger, or it may be a legitimate de- 
velopment of this, as eins is eens. — 
S.S.H. The latter hypothesis seems 
the more probable, and then the Eng- 
lish signification may have been at- 
tached to the German word from simi- 
larity of sound. Dr. Mombert thinks 
the word may be either any treated as 
a German word, or irgend einer cor- 
rupted. Observe the frequent use 
of (ee) for (ai) as eens for eins. The 
transitions of (an) into (aa), (ai) into 
{ee), (aa) into (oo), and ocasionally (o) 
in (u), are all noteworthy in connection 
with similar changes in English. 

39 Meaner for mehr is obscure. Com- 
pare Schmeller, Bayr. Wort. 2, 581 ; 
" manig, Schwab, menig, meng, a) wie 
hochd. manch .... Comparativisch 
steht in Amberg. Akten v. 1365 "An 
ainem stuck oder an mengern." . . . 
Sonst hdrt man im b. W. wie in 
Schwaben einfacher den Comparativ 
mener, mehr, welcher eher aus (mee, 
me) als aus menger ents'ellt scheint ; 
oder soUte es noch unmittelbar zum 
alten mana- gehoren?" 



F. "W. Gbseniub on the Language of Chauceh. 

Two German scholars, Professors Gesenius and Rapp, have pub- 
lished special studies on the langua<^e and pronunciation of Chaucer, 
of which it is now necessaiy to ^ve an account. The following is 
a condensed abstract of the treatise entitled : De Lingua Chauceri 
commentationem graramaticani scripsit Fridericus Guilelmus Ge- 
senius, Bonnae, 1847, 8vo. pp. 87. The writer (who must not 
be confounded with the late Prof. Wilhelm Gesenius, of Halle, the 
celebrated Hebraist,) used Tyrwhitt's text of the Canterbury Tales, 
according to the 1843 reprint. In the present abstract Wright's 
spelling and references to his ed. of Harl. MS. 7334 (which have 
all been verified) are substituted, and much relating to the pecu- 
liarities of Tyrwhitt's text is omitted ; inserted remarks are 
bracketed. Gesenius' s ags. orthography has been retained. 

Part I. The Letters. 

Chaucer seems to add or omit a final 
e at pleasure, both in ags. and fr. 
words, as was necessary to the metre ; 
and he used fr. words either with the 
fr. accent on the last syllable or with 
the present English accent, for the 
same reason. 

Chap. 1, Vowels derived from Anglo- 

Short vowels are followed by two 
consonants, or by either one or two in 
monosyllables, and long vowels have a 
single consonant followed by e final. 

I. Ags. short a is preserved in : land 
402, hand 401, bigan 5767, ran 4103, 
drank 6044, thanked 927 ; but fiuctu- 
ates often between a and o, as : londes 
14, hond 108, outsprong 13526. bygon 
7142, nat 2247, drank 13970, i-thanked 
7700 [in the three last cases, Tyrwhitt 
has o]. 

Short a answers to ags. a, according 
to Grimm's separation a = goth. a, 
and <e = gothic e, as: what, that pron., 
ags. hvat j^at ; atte. ags. at 29 ; glas 
152, have ags. habban, etc. 

Short a also answers to ags. ea, as 
in : alle ags. eall 1 0, scharpe ags. 
scearp 114, halle 372, barme 10945, 
starf 935, 4703, halpe [Tyrwhitt, hilp 
Wright] 5340, karf 9647, hals 4493. 

Long a is either a preserved ags. a 
long, or a produced ags. a short, as : 
make ags. macjan 4763, name, fare 
7016, ham, ags. ham 4030. That this 
last word was pronounced diff'erently 
to the others, which probably even 
then inclined to a (fe), is shewn by 
its interchange with home, whereas a 
always remains in make, name, etc. 

Long a also arises from ags. a short, 
as : smale ags. smal 9, bar 620 ; fadur 
100, blake 2980, this last vowel is 
sometimes short as 629. 

Long a like short a also arises from 
ags ea, as : gaf. ags. geaf 177, mary, 
ags. mearh 382, jape ags. geap 4341, 
ale 3820, gate 1895, care, etc. 

XL Chaucer's e replaces several dis- 
tinct ags. vowels. 

Short e stands 

for ags. e short, in : ende 15, wende 
16, bedde, selle 3819, etc. 

for ags. ^■, y, in : cherche (Wr. 
chirche), ags. circe 4987 ; selle ags. 
syl, threshold, 3820, rhyming with 
selle, ags. sylle ; scheeld ags. scyld 
2895, rhyming with heeld, ags. heold, 
kesse ags. cyssan 8933 ; stenten, ags. 
stintan 906 ; geven, ags. gifan, gyfan 
917, etc. These forms are only found 
when wanted for the rhyme, and i is 
the more common vowel. 

for ags. ea, ed in : erme, ags. ear- 
mjan 13727; erthe, ags. eard, eor'Se 
1898 ; ers, ags. ears 7272 ; derne, ags. 
dearn 3200, 3297 ; herd 272 ; est, ags. 
east 1905. 

for ags. ieo in : sterres, ags. steorra 
270 ; cherles ags. ceorl, ger. kerl, 
7788 ; yerne ags. georne, ger. gern, 
6575; lerne. ags. leornjan, 310; swerd 
112, werk 481, derkest 4724; yelwe, 
ags. geolu 677. * 

Long e stands 

for ags. short e in : ere, ags. eijan 
888 ; queen, ags. even 870, etc. 

for ags. long e, more frequently, in : 
seke, ags. secan 13 ; kene 104, grene 
103, swete 5, mete 1902, wepyng 2831, 
deme 1883. 

Chap. VII. § 1. 



for ags. ae long : heres, ags. haer 
557; breede, 1972; lere, ags. laeran 
6491 ; see 59, yeer 82, reed 3527, 
slepen 10, clene 369, speche 309, strete 
3823, etc. 

for ags. eo as in : seke, ags. seoc 18, 
as well as : sike, ags. sicca 245, these 
diphthongs eo, to, had probably a simi- 
lar pronunciation and are hence fre- 
quently confused, so heofon, Mo/on, 
and leo^, lio^ ; scheene, ags. sceone, 
beautiful, 1070 ; leef 1839, theef 3937 ; 
tene, ags. teona, grief, 3108; deepe 
129, chese 6480, tree 9337, tre 6341, 
prestes 164, prest 503, etc. 

for ags. ea and ed in : eek 5, gret 84, 
beteth 11078, neede 306, reede 1971, 
bene 9728, chepe 5850, deef 448, 
stremes 1497, teeres 2829, eet 13925, 
mere 544. 

Nothing certain can be concluded 
concerning the pronunciation of these 
^'s, which arose from so many sources. 
They all rhyme, and may have been 
the same. In modern spelling the e is 
now doubled, or more frequently re- 
verts to ea. 

III. The vowel i has generally re- 
mained unchanged at all periods of the 
language. Mention has already been 
made of its interchange with e where 
the ags y was the mutate of u or eo, to, 
thus: fist 6217, fest 14217, ags. fyst ; 
mylle 4113, melle 3921, ags. myll ; 
fel 5090, fille 10883, ags. feol ; develes 
7276, devyl 3901 [divel Tyrwhitt, 
deuel Heng. and Corp.], ags. dioful. 
The i generally replaces ags. y, and e 
replaces ags. eo. Long i similarly re- 
places long ags. y, as occasionally in 
ags. Short ags. ^ seems to have been 
lengthened before Id, nd, [no reasons 
are adduced,] as in: wylde 2311, 
chylde 2312, fynde 2415, bynde 2416. 
Undoubtedly this long z was then pro- 
nounced as now, namely as German 
ei (ai). [Pronunciatio longae vocalis 
I sine dubio iam id aetatis eadem fuit 
quam nunc, id est ei.^ In the con- 
tracted forms Jint, grint for findeth, 
grindeth, there was therefore a change 
of vowel, Jint having the German short 
t, and Jindeth German ei. [No reasons 

IV. Short stands 

for ags. short o in : wolde 651, 
god 1254. 

for ags, short u : somer ags. sumer 
396 ; wonne ags. wunnen 51 ; nonne 
118, Sonne 7, domb 776, dong 532, 
sondry, ags. sunder, 14, 25. Nearly 

all these words are now written with u, 
and preserve Chaucer's pronunciation, 
for summer is written, but sommer 
spoken [i.e. Gesenius did not distin- 
guish the sounds (a, o).] 

for ags. short a, as already observed, 
and is generally preferred before nd, 
and remains in Scotch and some 
northern dialects. 

Long stands 

for ags. long o in : bookes, ags. boe, 
1200 ; stooden 8981, stood 5435, took 
4430, foot 10219, sone 5023, sothely 
117, etc. 

for ags. long a in : wo, ags. va 8015, 
moo 111, owne, ags. agen 338, homly 
7425, on 31, goost 205, hoote 396, 
ooth 120, loth 488. In such words a 
is uncommon, the sole example noted 
being ham 4030. Both o's rhyme to- 
gether and were therefore pronounced 
alike. At present the first is u and the 
second o. 

for ags. short u in : sone 79 ; wone, 
ags. vunjan 337, groneth 7411. 

V. Short u stands for ags. short w 
in : ful, ags. full 90, lust 192, but 142, 
cursyng 663, uppon 700, sustcr 873, 
shulde probably arose from some form 
sculde, not sceolde, as we have no other 
instance of ags. eo becoming short u. 
There is no long u in Chaucer. 

VI. The vowel y is occasionally put 
for i. 

VII. The diphthong ay or ai stands 
for ags. iig in : day, ags. dag 19, weie 
793, lay 20, mayde 69, sayde 70, faire 
94, tayl 3876, nayles 2143, pleye 236, 
reyn 592, i-freyned, ags. fragnan 
12361. These examples shew that ey 
was occasionally written for ay, and 
hence that ey, ay must have been pro- 
nounced alike. 

VIII. The diphthong ey or ei arose 
from ags. ed as in : agein, ags. agean 
8642, or from edg as : eyen, ags. eage 
152, deye, ags. deagan 6802, \jnori, is 
there such a word in ags. ? it is not in 
Bosworth or Ettmiiller; Orrmin has 
deienn, supra p. 284. There is a 
deagan tingere.] The change in these 
two last words may be conceived thus : 
first g is added to ei, then replaced by 
j (j) and finally vanishes, as eige, eije, 
eie or eye. From eah comes eigh, as 
eahta, hedh, nedh, sledh, which give 
eyght, heygh, neygh, sleygh. This 
orthography is however rare, and highe, 
nighe, slighe, or hie nie slie, without 
gh, which was probably not pronounced 
at that time, are more common. The 




Chap. VII. § 1, 

word eight explains the origin of night, 
mighty etc., from ags, neaht, meaht, 
which were probably tirst written 
neight, meight, and then dropped the 
t. [There is no historical ground for 
this supposition.] 

IX. The diphthong oti, or oiv at the 
end of words or before e, answers to 
ags. long tc (as the German an to me- 
dieval German «), in : hour, ags. bur 
16153, oure 34, schowres 1, toun, ags. 
tun 217 ; rouned, ags. run 7132, doun, 
ags. dun 954 ; hous 252, oule 6663, bouk, 
ags. buce, Germ, bauch, 2748, souked 
8326, broukc, ags. brucan, use. 10182, 
etc. In many of these words ow is 
now written. 

Before Id and nd, ou stands sometimes 
for ags. short u. Before gh, ou arises 
from ags. long o, and answers to middle 
German uo, as : inough, ags. genog, 
mhg. genuoc 375 ; rought, ags. rohte 
8561, 3770, for which au is sometimes 
found, compare sale 4185, sowle 4261. 

Finally ou sometimes arises from 
ags. eov, as in : foure, ags. f cover 210 ; 
trouthe, ags. treovth, 46, etc. 

X. The diphthong eu, ew, will be 
treated under w. 

Chap. 2. Consonants derived from 


I. Liquids /, w, «, r. 

L is usually single at the end of 
words, though often doubled, as it is 
medially between a short and any 
vowel, but between a long vowel and 
a consonant it remains single. 

The metathesis of R which occurs 
euphonically in ags., is only found in : 
briddes 2931, 10925 ; thrid 2273, 
threttene 7841, thritty 14437 ; thurgh 
2619. But as these words have re- 
gained their primitive forms bird, 
third, through, we perceive that the 
metathesis was accidental. In other 
words the transposed ags. form disap- 
pears in Chaucer, thus : gothic rinnan, 
ags. irnan, Chaucer renne 3888 ; 
frankic drescan, ags. yerscan, Ch. 
threisshe 538, threisshfold 3482 ags. 
]?rescvold, ))erscvold ; frank, pr'estan, 
ags. berstan, Ch. berst [Harleian and 
Lansdowne bresten Ellesmere and 
Hengwurth, and Corpus, brestyn Cam- 
bridge,] 1982 ; goth. brinnan, ags. bir^ 
nan, Ch. breyi 2333 ; modern run, 
[urn in Devonshire], thrash, but burn 

II. Labials b, p, f, w. 

B is added euphonically to final m in 

lamb 4879, but not always, as lymei 
4881, now liynbs. 

P is used for b in nempnen 4927. 

F, which between two vowels was v 
in ags., is lost in heed 109, ags. hedfod, 
hedvod. There seems to be a similar 
elision of/ from ags. efenford in enforce 
2237 [emforth Ellesmere, Hengwrt, 
Corpus, enforte Cambridge, hensforth 
Petworth, enforlpe Lansdowne], com- 
pare han for haven 754, 1048, etc. jP 
is generally final, as : wif 447, lyf 
2259, gaf 1902, haf 2430, stryf 1836 
knyf 3958, more rarely medial, [the 
instances cited have final /in Wright], 
where it is generally replaced by v, 
not found ags., as: wyve 1862, lyves 
1720, geven 917, heven 2441, steven, 
ags. stefen 10464 ; havenes 409. 

V is never used finally, but is re- 
placed by w, followed sometimes by e, 
as : sawgh 2019, draw 2549, now 2266, 
sowe 2021, lowe 2025, knew 2070, 
bliew 10093, fewe 2107, newe 17291, 
trewe 17292. In the middle of a word 
aw, ow are replaced by au, ou, but 
before v, w is retained, as : howve 
3909, schowve 3910. 

^F arises from ags. g, as in : lawe, ags. 
lagu 311 ; dawes, ags. dag, 11492, and 
as day is more common for the last, we 
also find lay for the first, 4796. Com- 
pare also fawe ags. faegen 5802 rhym- 
ing with lawe, i-slawe 945, for fain, 
slain. W also replaces g in : sawe 
1528, 6241, mawe 4906, wawes 1960, 
sorw 10736, morwe 2493, borwe 10910, 
herberw 4143, herbergh 767, 11347. 

III. Linguals d, t, th, s. 

The rule of doubling medial conso- 
nants is neglected if Z) stands for ags. "S, 
as : thider 4564, whider 6968, gaderd, 
togeder, etc., in the preterits dide 
3421, 7073, 8739, and hade 556, 619, 
[Ellesmere and a few MSS. where it 
seems to have been an accommodation 
to the rhymes spade, blade.'] Similarly 
i-written 161, i-write 5086, although 
the vowel was short in ags. [It is 
lengthened by Bullokar in the xvith 
century, p. 114, 1. 7.] Perhaps litel 
has a long i in Chaucer's time, see 87, 
5254. ^ 

S final is often single, as : blis 4842, S 
glas 152, amys 17210.) 

The termination es in some adverbs 
is now ce, as : oones 3470, twyes 4346, 
thries 63, hennes hens 10972, 14102, 
henen 4031 [in Tyrwhitt, hey then 
Ellesmere, heithen Corpus, no cor- 
responding word in Harleian], henne 




2358 ; thennes 5463, 4930, thenne 
6723; whennes 12175. 

The aspirate TH had a double cha- 
racter ]? "S in ags., and a double sound, 
which probably prevailed in Chaucer's 
time, although scarcely recognized in 
writing. That th was used in both 
senses we see from : breeth, ags. brae's 
5 ; heeth, ags. hae'S 6 ; fetheres, ags. 
fe'Ser 107 ; forth, ags. for« 976 ; walk- 
eth 1054, etc. ; that, ags. j^aet 10 — 
ther 43, thanked 927. The use of 
medial and final d for th are traces of 
"S, as : mayde, ags. maeg'S 69 ; quod, 
ags. ova's 909 ; wheder ags. hvaSre 
4714 [whether, "Wright] ; cowde ags. 
cu'S 94 ; whether and cou])e are also 
found. Again, we also find [in some 
MSS.] the ags. d replaced by th, in : 
father 7937, gather 1055, wether, 
10366, mother 5433, [in all these cases 
Wright's edition has d']. But ^ on the 
other hand is never put for ags. \. 

The relation of th, s, is shewn by 
their flexional interchange in -eth, -es. 
The elision of th gives wher 7032, 

IV. Gutturals, c, k, ch, g, h, j, q, x. 
K is used before e, i, and c before 
a, 0, u, hence kerver 1801, kerveth 
17272, but: carf 100. Medial ags cc 
becomes cic or kk, as nekke, ags. hnecca 
238 ; thikke, ags. J^icca 551 ; lakketh 
2282, lokkes 679. Modern ck after a 
short vowel is sometimes k, as : seke 18, 
blake 2980. 

Grimm lays down the rule that c, k 
fall into ch before e, i except when 
these vowels are the mutates of a, o, w, 
in which cases k remains, (Gram. P, 
515.) cch has arisen from ags. cc in 
the same way as kk, as : wrecche, ags. 
vraecca 1 1332 fecche, ags. feccan 0942 ; 
cacche Mel., strecche, recche, etc. 
Probably the pronunciation was as the 
present tch. 

K was ejected from made, though 
the form maked remains 2526. In 
reule 173, if it is not derived from the 
French, the g of ags. regul, regol, has 
been ejected. 

G was probably always hard, and so 
may have been gg, in : brigge, ags. 
brycg 3920 ; eggyng ags. ecg, 10009 ; 
hegge, ags. hecg 16704. From this 
certainly did not much differ that gg 
which both in Chaucer and afterwards 
passed into ^■,as : ligge, lye ags. lecgan, 
2207; legge, ags. lecgan, 3935; abegge, 
abeye, ags, bycgan 3936. 

The g and y were often interchanged, 
as give yeve, forgete, forgate, gate yate, 
ayen agen, etc. The y replaced guttural 
g [due to editor] as in : yere, yonge, 
yerne, ey ; and also in words and ad- 
jectives where y arises from ig, as: 
peny, very, mery, etc., and in the pre- 
fix y or i for ags. ge, as : ylike, ynough, 
ywis, ymade, yslain, ywriten, ysene, 
ysowe 5653. And g we have seen is 
also interchanged with w. 

The hard sound of ags. h is evident 
from the change of niht, I'eoht, Jiiht, 
viht, etc., into night, light, flight, 
wight, etc. 

Ags. sc had always changed into sh, 
German sch. In some words ssh re- 
places sh as : fresshe, ags. fresc 90, 
wessch 2285, wissh 4873, asshy 2885. 
There is also the metathesis cs or x for 
sc in axe. 

Chap. 3. Vowel mutation, apocope, and 
junction of the negative particle. 

I. There is no proper vowel mutation 
{umlaut), but both the non-mutate and 
mutate forms, and sometimes one or the 
other, are occasionally preserved, as: 
sote 1, swete 5 ; grove 1637, greves 
1497, 1643 to rhyme with leves; wel- 
ken 9000, ags. wolcen. Germ, wolke ; 
the comparatives and superlatives, 
lenger, strenger, werst, aud plurals, men, 
feet, gees. 

II. Apocope; lite, fro, mo, tho = 

III. Negative junction; before a 
vowel : non = ne on, nother, neithir = 
ne other, ne either, nis=nQ is, nam = 
ne am ; before h ox w: nad = ne had, 
10212, nath = ne hath 925, m7=ne 
will 8522, nolde=ne wolde 552, nere 
=ne were 877, not = ne wot 286, 
nysten— ne wysten 10948. 

Chap. 4. Vowels derived from the 
French words with unaltered spelling 
were probably introduced by Chaucer 
himself, and the others had been pre- 
viously received and changed by popu- 
lar use. 

I. The vowel a in unaccented syl- 
lables had probably even then approxi- 
mated to e, and hence these two vowels 
are often confounded. Thus Chaucer's 
a replaces fr. e, ai, and again Ch. e re- 
places fr. a, thus : vasselage [see vas- 
selage, p. 642, col. 2, and wasseyllage, 
p. 645], fr. vasselage 3056, vilanye [see 
villany, p. 642, col. 2, and courtesy, 
p. 644, col. 1], fir. vilenie, vilainie, 



728 ; companye, fr. compaignie 4554, 
chesteyn [chastei/n, chestayn, in MSS., 
see p. 642,] fr. chastaigne 2924. 

With the interchange of the ags. 
vowels a, o, we may compare the change 
of fr. a, au, the latter having probably 
a rough sound as of ao united, which 
took place before wc, ns, ng, nd, nt in 
both languages, but au was more fre- 
quent in Chaucer and a in French, as : 
grevance 11253, grevaunce 15999, and 
other ance and ant terminations, also : 
roraauns, fr. romance 15305 ; en- 
haunsen, fr. enhanser 1436 ; straunge 
fr. estrange 10590, 10403, 10381; 
demaundes, fr. demande 8224 ; launde 
fr. lande, uncultivated district, 1693, 
1698 ; tyraunt, fr. tirant 9863, tyrant 
15589; graunted 6478, 6595; haunt 
fr. hante 449. With the exception of 
the last word all these have now a. 

II. Long e frequently arises from 
French ai, as in : plesaunce, fr. plai- 
sance 2487 ; appese, fr. apaisier 8309 ; 
freeltee, fr. fraiiete ; peere, fr. paire 
15540. Sometimes it replaces ie^ as : 
nece, fr. niez 14511 ; sege 939, siege 
56 ; and the e is even short in : cherte, 
fr. chierte 11193. Similarly fr. i is 
omitted in the infinitive termination 
t>r, compare arace, creance, darreine, 
auter, etc.. in the list of obsolete fr. 

Long e also replaces fr. eu in : peple 
2662 [the word is omitted in Harl., 
other MSS. have peple, poeple, puple], 
mebles [moeblis Harl.] 9188. To this 
we should refer : reproef 5598, ypreued 
[proved Harl., proeued Hengwrt] 487. 

III. That the pronunciation of i 
fluctuated between i and e we see by 
the frequent interchange of these let- 
ters ; the fr. shews e for It. i, as : de- 
vine 122, divyn 15543, divide 15676, 
divided 15720 [Tyr. has devide in the 
first case], enformed 10649, fr. in- 
former, enformer ; defame 8416, dif- 
fame 8606 ; surquidrie surquedrie, 
chivachee chevachie, see obsolete fr. 
words below. 

IV. Chaucer frequently writes o for 
fr. ou in accented syllables, as : cover- 
chefes [most MSS., hevercliefsYLdcA!] fr. 
couvrechief 455 ; corone, fr. couronne 
2292 ; bocler, fr. bonder 4017 ; govern- 
aunce, fr. gouvernance 10625 ; seve- 
re yn, fr. souverain 67. More rarely 
Ch. « = fr. ow, as : turne [most MSS., 
tourne Harl.], fr. tourner 2456 ; cur- 
tesye, fr. courtoisie 15982. 

V. Fr. is often replaced by Ch. «, 

as : turment [torment Harl.l, fr. tor- 
mente 5265 ; abundauntjy, fr. habon- 
dant 5290 ; purveans, fr. porveance, 
pourveance 1667; in aasuaye 11147, 
fr. assoager, assouager, the u had cer- 
tainly the sound of w, compare aswage 

For long u we occasionally find ewy 
wliich was certainly pronounced as in 
tlie present few, dew, thus : salewith 
[Harl. and the six MSS. read salueth'] 
1494, transmewed [translated Harl., 
transmeeuyd Univ. Cam. Dd. 4, 24] 826 
mewe, fr. mue 351 [muwe Ellesmere 
and Hengwrt MSS.] jewise, fr. juise 
[juwyse Harl. and most MSS., iwes 
Petworth, iuyse Lansd.] 1741. 

VI. The vowels y and i are inter- 
changed in fr. as in ags. words. 

VII. The fr. diphthongs ai, oi, 
usually appear as ei in Chaucer, and 
must have been pronounced identically, 
as: seynte, fr. saint 511; doseyn, fr. 
dosaine 580 ; chesteyn, fr. chastaigne 
2924 ; peyneth, fr. painer, peiner 4740 ; 
coveitous, fr. covoiteux, Mel. These 
diphthongs interchange in Ch. as well 
as in fr. [difi'erent MSS. differ so 
much that Gesenius's references to 
Tyrwhitt's edition on this point are 
worthless] . For the interchange of a 
and ai see I. 

VIII. WTien the diphthong ou arose 
from fr, 0, it was perhaps pronounced 
as long 0. This is very probable in 
those words which now contain o or u 
in place of the diphthong, but less so 
in those which have preserved ou ; as 
these had even then perhaps the sound 
of German au. Ex. noumbre 5607 ; 
facound, fr. faconde 13465, soun, fr. 
son 2434; abounde fr. habonder 16234. 
[The other examples have o in Wright's 
ed., or like Jlour 4 are not to the point; 
the above are now all nasal on.] 

Chap. 5. Consonants derived from the 

The doubling of final consonants is 
fi-equently neglected. 

I. Liquids. 

[The examples of doubling /, r, are 
so different in Wright's ed. that they 
cannot be cited.] 

F inserted : dampned 5530, damp- 
nacioun 6649; sompne 6929 =somone 
7159, sompnour 6909, solempne 209. 
This p is also often found in old fr. 
Similarly in Proven9al dampna, somp- 
nar, Diez. Gram. 1, 190 (ed. 1.). 



II. Labials. 

P for h\ gipser, fr. gibecier 359 ; 
capul, fr. cabal 7732. The letter v, 
which was adopted from the romance 
languages into English, had no doubt 
the same sound as at present, that is, 
it was the German w^ and the w was 
the German u. [That is, Ges. con- 
fuses (v, w) with (bh, u) in common 
with most Germans.] 

As in ags. g passes into German w, 
so in fr. words initial w becomes g or 
gu. Whether this change was made 
in English by the analogy of the ags. 
elements or from some other dialect of 
old fr., in which probably both forms 
were in use, it is difficult to determine. 
The following are examples : wiket, fr. 
guichet 10026 ; awayt, fr. aguet 7239 ; 
wardrobe, fr. garderobe 14983. To 
these appear to belong warice and 
wastear, though they may derive from 
the frankic warjan wastan. 

III. Lingual s. 

Z is an additional letter, but is sel- 
dom used, as lazer 242. Ch. generally 
writes s for z. 

IV. Gutturals. 

C before e, i was probably s as now. 
Fr. gn now pronounced as German nj^ 
(nj) is reduced to n in Ch., as Coloyne 
468, feyne 738, barreine, essoine, oine- 
ment. G was doubled after short 
vowels in imitation of ags. 

The aspirate A, which seems to have 
come from external sources into Eng- 
lish, and was scarcely heard in speech, 
was acknowledged by Ch., but has now 
disappeared, as : abhominaciouns 4508. 
In proheme 7919, the h seems only in- 
serted as a diaeresis. 

Fr. qu before e and i is often changed 
into k, as : phisik 913, magik 418, 
practike 5769, cliket 10025. 

Chap. 6. Aphceresis of unaccented 
French e, a. 
Initial e is frequently omitted before 
st^ sp, sc, as : stabled, fr. establir 2997; 
spices, fr. espece 3015; specially 14, 
squyer, fr. escuyer 79, scoler, fr. escolier 
262 ; straunge, fr. estrange 13. Similarly 
a, (9, are rejected in other words where 
they are now received, as : potecary 
14267, compare Italian bottega a shop ; 
prentis 14711, pistil 9030, compare 
Italian pistola, chicsa. The initial a 
in avysioun 16600, has been subse- 
quently rejected. 

Part II. Flexion. 

Chap. 1. On Nouns. 

Chap. 2. On Adjectives. 

Chap. 3. On Pronouns ^ Numerals. 

Chap. 4. On Verbs. 


I. Obsolete Chaucerian words of 
Anglosaxon origin. 

[All Gesenius's words are inserted, 
though some of them are still in fre- 
quent use, at least provincially, or have 
been recently revived. To all such 
words I have prefixed f. The italic 
word is Chaucer's, the roman word is 
ags., meanings and observations are in 
brackets. Gesenius seems to have sim- 
ply extracted this list from Tyrwhitt's 
Glossary without verification, as he has 
occasionally given a reference as if to 
Cant. Tales, which belongs to Eom. of 
Eose. The Mel. and Pers. T. refer to 
the tales of Melibeus and the Persoun, 
without any precise indication, as edi- 
tions differ so much.] 

abegge abycgan [abide] 3936, abeye 
13515, abye 12622 agrise agrisan 
frighten] 5034, algates algate algeats 
in any case] 573, 7619, anhang an- 
hangan [hang on] 13690, attry utterly 
atter atterlic Persons Tale [poisonous], 
awreke avrecan [wreak] 10768. 

bale [p. 379], barme bearm [lap] 
10945, bedred beddredda [bedridden] 
7351, 9168 ; biknowe becnavan [con- 
fess] 5306, blynne blinnan [cease] 13099, 
blyve [quickly, supra p. 380, col. 2], 
borive [supra p. 380, col. 2 ; where for 
loan read security']., bouk buce [belly] 
2748, byleve frank, pilipan, germ, blei- 
ben, [remain] 10897. 

fchaffare ceap + faran ? germ, kauf- 
fahren [chaffer, bargain] 4558, clepe 
clypjan [call] 3432, [name] 121, etc., 
colde [to turn cold] 5299, ■fcop cop 
[top] 55Q, ^<?/dofjan [daft] 4206, dere 
derjan [hurt] 1824, 10554, derne dearn 
dyrn [hidden p. 382] 3278, 3297, 
dighten dihtan [dispose] 6349, 16015, 
■fdomesman [judge] 15976. 

eft aft eft [again] 1671, 5212, eft- 
sones [soon again] 6390, eftsoone 16082, 
■\eek eac [eke] 5, \elde yldo eldo [old 
age] 6797, emforth [supra p. 666, col. 2, 
1. 8,] ^ere erjan [to plough] 888, erme 
earmjan [to pity] 13727, ers, ears ars 
[arse] 3732, 7276. 

fele fela feola [many] 8793, fere 
[companionship, supra p. 383], ^fit fitt 
[song] 15296, fleme aflyman [drive 
away] 17114,/o floga? [arrow] 17196, 



Chap. VII. ^ 1. 

fonge fangan [take] 4797, forpine 
pinan [waste away] 205, forward foro- 
veard [promise] 831, 850, 854, 4460, 
/r<jy«« gefregnan [ask] \2^^\, fremde 
fremed [strange] 10743. 

gale galan [yell] 6414, 6918, fgar 

fearvan [make ; the word is get in 
[arl., Heng., Corp., gar in Tyrwhitt' 
4130, girden geard gyrd? [cut off" 
16032, gleede gled [heat] 3379, gnide 
gnidan [so Tyr., girdyng Harl., gig- 
gynge EUes., Cam., gyggynge Heng., 
gydyng Corp. gideing Lans., sigyng 
Pet.] 2504, grame grama, ger. gram 
[grief] 13331, greyth hra'Sjan [pre- 
pare] 4307, graithe 16080. 

hah heals [neck] 4493, halse heals- 
jan [embrace] 15056, [heende frank, 
pihandi, germ, behende [swift ? cour- 
teous, supra p. 385] 3199, 6868, hente 
gehentan [to take] 700, hent 7082, 
herde hirde [shepherd] 605, 12120, 
herie herjan [praise] 5292, 8492, heste 
haes [command] 14055, by heste 4461, 
heete [promised] 2400, hete 4754, ■fhight 
[call] 1015, fhie higan, on hye [in 
haste] 2981, in hyghe [in haste] 4629. 
hine hina [hind p. 385] 605, ■fholt 
holt, germ, holz [wood] 6. 

jape geap [joke] 707, 4341, 13240, 
[to joke] 15104. 

kithe cySan [announce] 7191, keked 
germ, gucken [Corp., loked Harl., liked 
Heng.] 3445, latered [delayed] Pers. 
Tale, \leche laece 3902, lydne lyden 
[language] 10749, leemes leoma [ray : 
beemes Harl.] 16416, lere laeran [teach] 
6491, 10002, levene [lightning] lige ? 
more probably than, hlifjan 5858, 
flewed laevd leaved [ignorant] 6928, 
7590, lissed lysan [loosed] 11482, [re- 
mission] 11550, nth li^ [limb] 16361, 
lit her ly lySr la's [bad], ger. liederlich, 

make maga mag, [husband] 5667, 
[wife] 9698, [match] 2558. 

nempnen nemnan nemjan [name] 
4927, note notu [business] 4066. 

oned [united] 7550. 

■\pan panne [brainpan, skull] 15438. 

rathe bra's bra's [quick] 14510, 
freeche recan [reck, care] 2247, 4514, 
reed raed [advice] 3527, [to advise] 
3073, reyse goth. urraisjan [travel] 54, 
rys arisan, germ, reisholz [twig] 3324, 
rotme run 7132, rowne 10530, rode 
rude [ruddiness, face] 3317, 15138. 

fsawe sagu [saying] 1528, schawe 
scuva scua [shade, grove] 4365, 6968, 
shymeryng sciman scimjan, ger. schim- 
mern, [Heng., glymeryng Harl.] 4295, 

acheene seine sceone scone, ger. schbn 
[beautiful] 1070, 10202, f.shepen scy- 
pcn, ger. schoppen [stable] 6453, 
Hchonde sceonde [disgrace] 15316, 
fsibbe sib [relation] Mel., sikurly 
frank, sihhur, germ, sicher 137, secur 
[ib.] 9582, sithe si^ [times] 5575, 5153, 
sithen sith sin si'S'San 4478, 1817, seth 
5234, schenchith scencan [pour out 
wine] 9596, smythe smi'San [forge 
3760, sonde sand [message, messenger' 
4808, 14630, fsparre sparran [spar' 
992, star/ staerf [died] 935, 4703, 
Steven stefen [voice] 10464, stounde 
stund [space or time] 3990, fstreen 
stre6nan [parents] 8033, sw^/^^ sveltan 
"die] 3703, swelde 1358, sweven svefen 
'dream] 16408, etc., swithe svi'S 
'quickly] 5057. 

ftene teona [loss] 3108, thewes J^e^v 
[morals] 8285, tholid ]?61jan [suffer] 
7128, ■\threpe j^reapjan [blame] 12754, 
twynne tvinjan tveonjan [doubt, sepa- 
rate] 837, 13845. 

unethe ea'Se [uneasily] 3123, unhele 
unhaelu [affliction] 13531, unright un- 
riht [injury] 6675. 

wanhope vanjan + hopa [despair] 
1251, Wig/^^e'i^ vlacjan? frank, welchon, 
germ, verwelkt [withered] 14153, 
fwelken volcen 9000, [Harl. reads 
heven 16217, Tyr. welken], fwende 
[went] 21, whil er [shortly, just now] 
13256, ftvhiloni hvilura, ger. weiland 
861, wisse visan [shew] 6590, wone 
vunjan [dwell] 337, fwood vod [mad] 
1331, woodith [rageth] 12395. 

yerne georne 6575, -fyede eode [went] 
13069, ywys gewis [certainly] 6040. 

II. Obsolete Chaucerian words of 
French origin. 

[The italic word is Chaucer's, the 
roman the old French as given by 
Gesenius on the authority of Eoquefort ; 
when this is not added the word was 
unchanged by Chaucer, Meanings and 
remarks are in brackets. This list again 
contains many words not really obso- 
lete, here marked with f.] 

agregge agregier [aggravate] Mel., 
amoneste [admonish] Mel., anientissed 
anientir [annihilated] Mel., arace ar- 
rachier [tear] 8979, -far ray, [order] 
8138, [state, condition] 718, 8841, 
4719, [dress] 8860, [escort] 8821, [to 
put in order] 8837, arette arester [ac- 
cuse, impute] 726 [Harl., Corp., Pet., 
Lans., have ret, rette, the others na- 
rette\ 2731, \assoile [solve, absolve] 
9528, attempre attemprer 16324, Mel., 



avaunte avanter [boast] 5985, avaun- 
tour [boaster] Mel., avoutrie [adultery] 
6888, advoutrie 9309, auter autier 2294, 
awayt aguet [watch] 7241, 16211, 
ay el aiel [grandfather] \ayel Harl., 
ay ell Corp., Lans., aiel Elles, Heng. 
Cam., eile Pet.] 2479. 

■fbareigne baraigne [barren] 8324, 
bareyn 1979, ^haudery bauderie [joy] 
1928, -fbenesoun beneison 9239, blandise 
blandir Pers. T., bobaunce boubance 
6151, borel burel [rough dark dress] 
5938, [rough] 11028, bribe [broken 
meat after a meal] 6960, [beg] 4415, 
burned hMxmx 1985. 

cantel [fragment] 3010, -fcatel catels 
[goods] 542, 4447, fcharbocle [carbun- 
cle] 15279, chesteyn chastaigne [chest- 
nut] 2924, chivachie chevauchee [ca- 
valry expedition] 85, cJiivache 16982, 
clergeoun clergeon [acolyte] 14914, 
corrumpable [corruptible] 3012, costage 
[cost] 5831, covine [practice, cunning] 
606, coiclpe [fault] Pers. T., custumance 
custom] 15997, creaunce creancier 
^act on credit] 14700, 14714. 

dereyne derainier [prove justness of 
claim] 1611, 1633, delyver delivre 
[quick] 84, f disarray desarray [con- 
fiision] Pers. T., dispatisoun disputison 
[dispute] 11202, dole dol [grief, no re- 
ference given, 4*38], drewery druerie 
[fidelity] 15303. 

egrimoigne agrimoine [agrimony] 
12728, enchesoim enchaison [cause] 
10770, engendrure [generation] 5716, 
engregge engreger [aggravate] Pers. T., 
enJiorte enhorter [exhort] 2853, fentent 
[intention] 3173, fesehue eschuir 
[avoid] Mel., essoine essoigne [excuse] 
Pers. T., estres [situation, plan of 
house] 1973, 4293. 

faiteur faiteor [idle fellow, no re- 
ference], false falser [to falsify] 3175, 
ffey fee [faith] 3284, ffers [fierce] 
1600, fetys [beautiful] 157, fiaunce 
fiance [trust, false reference, 6" 167] 
fortune fortuner [render prosperous] 

garget gargate [neck] 16821, ^gent 
[genteel] 3234, gyn engin [trick] 10442, 
13093, (7iY<?rwi;gisterne guiterne [guitar] 
3333, 4394, gonfenon [standard 6*62, 
gounfaucoun 6*37] . 

fharie harier [persecute] 2728 [rent 
Wr., haried, the Six MSS.], herburgage 
[dwelling] 4327, humblesse [humble- 
ness] 4585. 

jambeux [leggings] 15283, jangle 
jangler [to jest] 10534, [a jest] 6989, 

juwise juise [judgment] 1741, irous 
ireux [angry] 7598. 

lachesse [negligence] Pers. T., letua- 
ries [electuaries] 428, 9683, letterure 
lettreure [literature] 15982, 12774, 
loos los [praise, good fame] 13296, 
Mel., losengour [flatterer] 16812. 

Mahoun Mahon [Mahomet] 4644, 
fmaistrie [master's skill] 3383, [mas- 
tery] 6622, 9048, f malison maleiceon 
[malediction] Pers. T., fmanace ma- 
nacher [menace] 9626, maat mat [sad] 
957, matrimoigne [matrimony] 9447, 
maumet mahommet [idol] Pers. T., 
merciable [merciful] 15099, mesel 
[leper] Pers. T., meselrie [leprosy] Pers. 
T., fmewevoMQ [place for keeping birds] 
351, 10957, mester [mystery, business, 
trade] 615, 1342 [except in Harl., 
which reads cheer. 1 

nakers nacaires [kettledrums] 2513, 
nyce [foolish] 6520, nycete 4044. 

foynement oignement 633, olifaunt 
olifant [elephant] 15219, op ye [opium] 

f palmer palmier 13, parage [parent- 
age] 5832, parjight parfyt parfit [per- 
fect] 72, 3011, parte parter [take part 
in] 9504, f penance [penitence] Pers. 
T., [penance] 223, [affliction] 5224, 
11052, pcnant [penitent] 15420, po- 
raille [poor people] 247, prow prou 
[profit] 13715, fpurveance pourveance 
[providence, forethought] 1254, 6152, 
3566, piiterie [whoredom] Pers. T., 
putour [whoremonger] Pers. T. 

rage ragier [sport] 3273, real [royal] 
15630, rially [royally] 380, reneye 
reneier [renounce] 4760, 4796, repeire 
[retm-n] 10903, respite 11886, \route 
[crowd] ger. rotte, 624. 

fsolas [joy, pleasure] 800, 3654, 
sourde sourdre [to rise] Pers. T., sur- 
quedrie [presumption] Pers. T. 

talent [inclination, desire] 5557, Pers. 
T. tester testiere [horse's head armour] 
2501, textuel [texted ivel Wr., having 
a power of citing texts] 17167, trans- 
m^^^(? transmuer [translated Wx,'] 8261, 
tretys traictis [well made, streight Wr.] 
152, ftriacle [remedy] 4899, trine trin 
[triune] 11973. 

vassclage [bravery] 3056, fverray 
[true] 6786, fversijlour versifieur 
[versifyer] Mel., viage veage [journey] 
77, 4:679, fv it ai lie [victuals] 3551, void 
voider [to remove] 8786, [to depart] 
11462, [to leave, make empty] 9689. 

war ice garir [heal] 12840, [grow 
whole], Mel. fwastour gasteur [waster] 



Chap. VII. § 1. 

M. Rapp on the Pronunciation of Chaucer. 

Dr. Moritz Rapp, at the conclusion of his Vergleichende Gram- 
matik, vol. 3, pp. 166-179, has given his opinion concerning the 
pronunciation of Chaucer, chiefly on a. priori grounds, using Wright's 
edition, and has appended a phonetic transcription of the opening 
lines of the Canterbury Tales as a specimen. This account is here 
annexed, slightly abridged, with the phonetic spelling transliterated 
into palacotype, preserving all the peculiarities of the original, such 
as absence of accent mark, duplication of consonants, German (bh) 
for (w), modern English errors of pronunciation, etc. A few re- 
marks are added in brackets. 

The liquids are to be pronounced as 
written, and hence I is not mute, 
though there is a trace of its disap- 
pearance in the form (naf) for (H«lf). 
The transposition of r is not complete ; 
we again find (rcnnc) for (irnrm), and 
(brenne) for (birnrm), English (rann, 
bsrn), (thurkh) through is unchanged, 
(bird) and (brid) are both used, 
( thresh e) replaces (therskan), and 
(breste) replaces (berstan), JEnglish 
(b^rst) . 

Among the labials, b remains after 
m in (l«mb), but (limm) is without the 
present mute b. For (nemnrm) we 
nave the peculiar (nempnen), and 
similarly (d«mpnen) to damn. Final 
/ as in (bhiif ) wife, is also written 
medially wive, that is, in the French 
fashion, because v tended towards / in 
the middle ages. But initially, in 
order to preserve the pure German (bh), 
recourse was had to the reduplication 
uu or w. On w after a vowel see 
below. (Bh) sometimes arises from a 
guttural, as sorwe, that is, (sorbhe) 
now sorrow = (sdrroo), from soi^g. 

Among the dentals d and t occasion 
no difficulty, and s has, by French in- 
fluence, become pure (s), [Dr. Rapp 
holds it to have been (sj) in ags.] 
especially as it sometimes results from 
]?. The z is merely an s. The most 
difficult point is th. In ags., we have 
shewn [supra p. 555, note] that it had 
only one value (th). I consider that 
this is also the case for this dialect. 
As regards the initial sound, which in 
the English pronouns is (dh), there is 
not only no proof of this softening, but 
the contrary results from v. 12589 

So faren we, if I schal say the sothe. 

Now, quod oure ost, yit let me talke 
to the. 
The form sothe has here assumed a 
false French e, since the ags. is (sooth) 

and English (suuth), [it may be the 
adverbial e, or the definite e, according 
as the is taken as the pronoun or the 
definite article,] which must therefore 
have here been called (soothe), as this 
th is always hard, and as to the, i.e. 
(too th^^) rhymes with it, shewing that 
the e of sothe was audible if not long, 
and that the th of to the was neces- 
sarily hard, as the English (tuu dhii) 
would have been no rhyme, [but see 
supra p. 318]. Similar rhymes are 
(aluu thee) allow thee, and (juuthe) 
youth, (nil thee) hie thee, and (sbhiithe) 
quickly, [supra pp. 318, 444, n. 2]. The 
Anglosaxon value of the letters must 
be presumed until there is an evident 
sign of some change having occurred. 
For the medial English th we have a 
distinct testimony that the Icelandic 
and Danish softening of d into (dh) 
had not yet occurred, for the best MSS. 
retain the ags. d, thus : ags. (feeder) 
here {fader), now (faadher), (gaderjan) 
here {gadev) now(g<Tedhdhar),(tog8edere) 
here (tognder) now (togiidhdhar), (bhE- 
der) here (bhEder) now (uEdhdhar), 
weather, (moodor) here (mooder) now 
(madhdhar) mother, (khbhider) here 
(khbhider) now (huidhdhar) whither, 
(thider) here (thider) now (dhidhdhar) 
thither. Inferior MS. have father, 
gather, thither, etc., shewing that the 
softening of d into the Danish (dh) 
began soon after Chaucer. But when 
we find the d in Chaucer it follows as 
a matter of course that the genuine 
old \ (th) as in (broother, fether) when 
here written brother, fether, could only 
have had the sound (th), and could 
not have been pronounced like the 
(bradhdhar, fEdhdhar). The ags. hu^e 
is here (kuth) and also (kud) or (kuud) 
for (kun-de.) 

Among the gutturals, h is written 
for c when e or * follows, and before 

Chap. VII. § 1. 



n as (knEu) knew. The reduplicated 
form is ck. The g is pure (g) in the 
German words, but in French words 
the syllables ffe, gi, have the Provencal 
sounds (dzhe, dzhi), which is certainly 
beyond the known range of Norman or 
old French, where g is resolved into 
simple (zh), but here gentil is still 
(dzhentil) not (zhentil). Similarly 
romanic ch is (tsh), and this value 
is applied to old naturalised words, 
in which the hiss has arisen from 
^, as (tshertsh) from (kirk), (tshcfp) 
from (keapjA-n) cheapen, and in 
thoroughly German words (tshild 
from (kild) child ; and (aelk) be- 
comes (^<;tsh) each. Reduplication is 
expressed by cch, representing the 
sharpened (tsh) [i.e. which shortens the 
preceding vowel] so that (bhrsekka) 
exile becomes wrecche, and sometimes 
wretch, which can only mean (bhrEtsh) ; 
similarly from (fekkr/n) comes (fetshe) 
and in the same way (retshe, stretshe) 
and the obscure cacche = (krttshe), 
which comes from the Norman cachier, 
although (tshase) also occurs from the 
French chasser. The reduplicated g 
occasions some difficulty. In French 
words abbrcgier can only give abregge 
= («bredzhe), and loger gives (lodzhe), 
etc., but the hiss is not so certain in 
brigge bridge, egge edge, point, hegge 
hedge, as now prevalent, because we 
find also ligge and lie from (liggfl'n) 
now (lai), legge and (lEEie) from (leg- 
gan) now (1^^), and (abEEie) from 
(byggan) now (bai). Similarly (bEgge) 
ask, beg, now (bEg), which, as I be- 
lieve, was formed from (buug«n) or 
(bEgefl-n) to bow. Here we find mo- 
dern (dzh) and hence the (dzh) of the 
former cases is doubtful. 

The softening of g into (j) is a 
slighter difference. The letter (j) does 
not occur in ags., and has been replaced 
in an uncertain way by ^■, g, ge. In 
Chaucer the simple sign y is employed 
[more generally j, the y is due to the 
ectitor, p. 310], which often goes fur- 
ther than in English, as we have not 
only {seev) a year, but give and (jEve, 
j«f, forjEte, j«t, ajEn, ajEust) and (ee) 
or (eeI) an e^g. 

The termination ig drops its g, as 
(pEni) for penig, and the particle ge 
assumes the form i, as (inuukh) enough, 
(ibhis-) certain, and in the participles 
(itflken) taken, (imAAd) made, (IsIaa) 
or (islEEn) slain, (is^ene) seen, (ibhriten) 
written, etc. From (geliike) comes 

(iliik) or (iliitsh), and the suffixed 
(-liik) is reduced to (li). 

The old pronunciation (qg) must be 
retained for ng, thus (loqg, loqger) or 
(leqger) ; there is no certain evidence 
for (loqq). The French nasal is in pre- 
ference expressed by n. "What the 
Frenchman wrote raison and pro- 
nounced (rEEsoq*) is here written resoun 
and called (resuun), as if the (q) were 
unknown. As the termination in 
givende has assumed the form {giving), 
we might conjecture the sound to be 
(giviq), because the form comes direct 
from (givin), as the Scotch and com- 
mon people still say, but we must re- 
member that giving also answers to the 
German Gebung, in which the g is 

We now come to h, which is also 
a difficulty. That initial h before a 
vowel had now become (h') as in Ger- 
man of the XIII th century, is very pro- 
bable, because h was also written in 
Latin and French words, and is still 
spoken. Chaucer has occasionally 
elided the silent e in the French fashion 
before A, which was certainly an error 
\ivas freilich ein Missgriff ivar ! 
shared by Orrmin, supra p. 490, and 
intermediate writers, who were free 
from French influence.] For the me- 
dial //, the dialect perceived its differ- 
ence from (h'), and hence used the new 
combination gh, known in the old 
Flemish, where the soft (kh) has been 
developed from g. The ags. niht-= 
(nikht) became night = (nikht), and 
similarly thurgh — (thurkh). For 
(khlEakhan) we have laivh, and 
laugh, both =(lAAkh); (sEakh) gives 
sawh — (sAAkh) or seigh = (sEEkh). 
Before I, n, r, the ags. h has disap- 
peared, but ags. (khbhiite) is here 
somewhat singularly written white, a 
transposition of hwite. Had k been 
silent it would have been omitted as in 
hi, hn^ hr, but as it was different from 
an ordinary h before a vowel, this ab- 
normal sign for (khbh), formed on the 
analogy of gh, came into use, and 
really signified an abbreviated heavy 
ghiv. Hence (khbhiite) retained its 
Anglosaxon sound in Chaucer's time. 
[Rapp could not distinguish English w 
from (u), and hence to him ivh was 
(hu), the real meaning of wh thus 
escaped him. His theory is that h 
was always (kh) in the old Teutonic 

We have still to consider sk and hs. 



Chap. VII. $ 1. 

The former was softened to (sjkj) in 
ags., and hence prepared the way for 
the simple (sh), and this may have 
nearly occurred by Chaucer's time, as 
he writes sch which bears the same re- 
lation to the French c/i = (tsh), as the 
Italian sci to ci^ s shewing the omission 
of the initial t. Some MSS. use ssh 
and even the present aA, the guttural 
being entirely forgotten. The ags. A\s- 
remains, but sk is still transposed into 
ks in the bad old way, as fla;^ = (akse) 
for (ffske). 

For the vowels, Gesenius has come 
to conclusions, which are partly based 
on Grimm's Grammar, and partly due 
to his having been preoccupied with 
modern English, and have no firm 
foundation. The Englishmen of the 
present day have no more idea how to 
read their own old language, than the 
Frenchmen theirs. We Germans are 
less prejudiced in these matters, and 
can judge more freely. Two conditions 
are necessary for reading old English 
correctly— first, to read Anglosaxon 
correctly, whence the dialect arose ; 
secondly, to read old French correctly, 
on whose orthography the old English 
was quite unmistakably modelled. 
[The complete catena of old English 
writers now known, renders this asser- 
tion more than doubtful. See supra 
p. 588, n. 2, and p. 640.] 

"We must presume that the old 
French a was pure (a). The ags. a, 
was lower = (a). The English ortho- 
graphy paid no attention to this diifer- 
ence, and hence spoke French a as {a). 
There can be no doubt of this, if we 
observe that this a was lengthened into 
au or aw, the value of which from a 
French point of view was (aa), as it 
still is in English, as straunge, de- 
maunde^ tyraunt, graunte, haimte. In 
all these cases the Englishman en- 
deavours to imitate French nasality by 
the combination (aau). [This au for 
a only occurs before n, see supra p. 
143, and infra Chap. VIII., § 3]. 

The old short vowel a hence remains 
{a) as in ags, thus (makjan) is in the 
oldest documents (makie, maki) and 
afterwards (make), where the (a) need 
no more be prolonged by the accent 
than in the German machen (makh'(?n), 
and we may read (makke). [But see 
Orrmin's makenn, p. 492]. 

The most important point is that the 
ags. false diphthongs are again over- 
come ; instead of (Ealle) we have the 

older form (alle),in.stead of (skKarp) we 
find (shrtrpe) etc. The nasal (an), aa 
in ags., is disposed to fall into (ou), as 
(h&nd, I'^nd, df(y(ik, begonne), etc. 

The greatest doubt might arise from 
the ags. (B or rather (ui) appearing as 
(«) without mutation; thus, ags. (thajt, 
khbhiet, bhajter, smail) again fall into 
(thrtt, khbh«t, bh«ter, sm«l). The mu- 
tation is revoked — that means, the ags. 
mutation had prevailed in literature, but 
not with the whole mass of the people, 
and hence in the present popular for- 
mation might revert to the older sound, 
for it is undeniable that although the 
present Englishman says (dha)t) with 
a mutated a, he pronounces (nuat, 
UAAtar, smAAl) what, water, small, 
without a mutate. In most cases the 
non-mutated form may be explained by 
a flexion, for if (doeg) in ags. gave the 
plural (dagrts), we may understand how 
Chaucer writes at one time (dEE) day 
and at another (dAA) daw for day, 

Short e remains unchanged as (e) 
under the accent, when unaccented it 
had perhaps become (a). Even in ags. 
it interchanges with i, y, as (tshirtsh) 
or (tshertsh) church. The ags. eo is 
again overcome, for although forms like 
beo^ beo]>, still occur in the oldest monu- 
ments, e is the later form, so that 
(stEorr«) star again becomes (stsrre), 
and (gEolu) yellow gives (jElbhe, jeIu), 
(fEol) fell becomes (fsll, fill), etc. A 
short (e) sometimes rhymes with a long 
one in Chaucer, as (mede, rij^de) mea- 
dow, red. Such false rhymes are how- 
ever found in German poetry of the 
XIII th century, and they are far from 
justifying us in introducing the modem 
long vowel into such words as (make, 
mEde), etc. 

The old long vowel e is here {ee)y as 
appears all the more certainly from its 
not being distinguished in writing from 
the short. [Eapp writes e e, but he 
usually pairs e e, a e = (e^ e, ee e), the 
(ee) being doubtful, {ee, ee). This 
arises from German habits, but in 
reality in closed syllables (e) is more 
frequent than (e), if a distinction has 
to be made. It would perhaps have 
represented Rapp more correctly to 
have written {ee e, ee e), but I con- 
sidered myself bound to the other dis- 
tribution, although it leads here to the 
absurdity of making {ee, e) a pair]. 
The quantity of the ags. must be re- 
tained, hence {seekan, k^me) can only 
give {seeke, keeii) seek, keen, and from 


Chap. VII. § 1. 



(sbh^ete) we also obtain (soote), with 
omitted (ee), compare Norse (soeoet) 
sweet. [The careful notation of quan- 
tity by Orrmin points him out as a 
better authority for this later period.] 
Long {ee) also replaces ags. ce as i^eexQ, 
see, skepe) hare, sea, sleep, and the old 
long eo as [seeke, leefe lee\e, d^(?pe, 
tsh^ese) seek, lief, deep, choose, and 
finally the old long ea as (eek) from 
(eak), and similarly {greete, b^me, 
tsh(?<?pe) great, bean, cheapen. These 
different {ee) rhyme together and have 
regularly become (ii) in modern Eng- 
lish. There is no doubt about short 
i, and long i could not have been a 
diphthong, because the French ortho- 
graphy had no suspicion of such a 
sound, Ags. y is sometimes rendered 
by ui as ftcire fire, which, however, 
already rhymes with (miire) and must 
therefore have sounded (fiire). The 
(yy) had become (ii) even in ags., so 
that (bruud) becomes (briide), etc. 
Least of all can we suppose short ^ in 
(bhilde, tshilde, finde) wild, child, find, 
to be diphthongal, or even long, as the 
orthography would have otherwise been 
quite difi'erent. 

Short may retain its natural sound 
(o), and often replaces ags. u, thus 
(sumrtr) gives (sommer), and (khnut, 
fiirthor) give (not, forther) nut, further. 
In these cases the Englishman gene- 
rally recurs to the mutate of (u), to be 
presently mentioned. 

Long in Chaucer unites two old 
long vowels, (aa) in (Hoome), some- 
times (HAm), (goost from (gAAst), 
(oothe) from (AAth) oath, (noote) from 
(HAt) ; and the old (oo) in (booke, 
tooke, foote, soothe). Both {oo) rhyme 
together, and must have, therefore, 
closely resembled each other ; they can 
scarcely have been the same, as they 
afterwards separated ; the latter may 
have inclined to (u) and has become 
quite (u). 

The sound of (u) is in the French 
fashion constantly denoted by o^«. [But 
see supra p. 425, 1. 3. Bapp is pro- 
bably wrong in attributing the intro- 
duction to French influence.] French 
raison was written raisun by the Anglo- 
Norman, and resoun by Chaucer, which 
could have only sounded (resuun) . A 
diphthong is impossible, as the name 
Cawcasous Caucasus rhymes with hous, 
and resoun with toun. Hence the 
sound must have been (huus, tuun) as 
in all German dialects of this date. 

Hence we have (fluur) flower for the 
French (floeoer). The real difiiculty 
consists in determining the quantity of 
the vowel, as it is not shewn by the 
spelling. Position would require a 
short (u) in cases like (shulder, bund, 
stund, bunden) shoulder, old (skulder), 
hound, hour, bound ; but the old 
(sookhte) must produce a (suukhte) 
sought ; and cases like (brukhte, 
thukhte) brought, thought, are doubt- 

On the other hand the vowel written. 
w, must have been the mutate common 
to the French, Icelander, Dutchman, 
Swede. The true sound is therefore 
an intermediate, which may have fluc- 
tuated between (ce, u, y), (lyst, kjTs) 
desire, curse. These u generally de- 
rive from ags. u, not y. The use of 
this sound in the unaccented syllable is 
remarkable. The ags. (bathjan) has two 
forms of the participle (bathod, bathed). 
Hence the two forms in Chaucer, 
(bflthyd) or rather (bathud) exactly as 
in Icelandic [where the ?^ = (-?), not (u), 
supra p. 548], the second (bathid, 
b«thed). Later English, however, 
could not flx this intermediate sound, 
and hence, forced by the mutations, gave 
the short u the colourless natural vowel 
(a), except before r where we still hear 
(^), [meaning, perhaps (ao). This theo- 
retical account does not seem to re- 
present the facts of the case.] The 
above value of short (u) in old Eng- 
lish is proved by all French words 
having this orthography. Sometimes 
Chaucer endeavours to express long 
(yy) by ui, as fruit, where, however, 
we may suspect the French diphthong ; 
but generally he writes nature for 
(natyyre) without symbolising the 
length. "We should not be misled by 
the retention of the pure (u) in mo- 
dern English for a few of these mu- 
tated u, as (full, putt, shudd, fruut). 
These anomalies establish no more 
against the clear rule than the few pure 
(a) of modern English prove anything 
against its ancient value. 

The written diphthongs cause pecu- 
liar difficulties. The combinations ai, 
ay, ei, ey, must have their French 
sound (ee), but as they often arise 
from (seg) there seems to have been an 
intermediate half-diphthongal or triph- 
thongal (eeI) ; thus (dsege) gives (dsEi) 
or (dEE). From eage) we have the 
variants eye, ye, eiyhe, yghe, so that 
the sound varies as {eeiQ, iije, iie, 



Chap. VII. § 1. 

Eikhe, iikhe). Similarly (niikhe) and 
(niie) liio;h, and (nEEkhe, niie) nigh. 
"We have already considered au, aiv, to 
have been (aa). The ags. (l«gu, lakh) 
law, gives luive, which perhaps bor- 
dered on a triphthongal (Iaauc). In 
the same way we occasionally find 
(dAAue) day, in two syllables, instead 
of the usual (dEE), ags. (daeg, dwgrts), 
and from ags. (sAAbhl) comes saule = 
(sAAle) and souie, which could have 
only been (suule). The medial otv = 
ou, that is, (uu), but before a vowel it 
might also border on a triphthong ; 
thus loiv/i = (luukh) low, is also written 
lotve = (looue) ? Oughen = (uukhen), 
and also owen = (oouen), now own = 
{oon). Similarly growe may have 
varied between (gruue, grooue) and so 
on with many others. These cases 
give most room for doubt, and the 
dialect was probably unsettled. Eut 
the diphthong exi, exv, leaves no room 
for doubt ; it cannot be French (oe) 
for hetire hour is here (nyyTe) [proba- 
bly a misprint for (nuure)], and for 
peuple we also find (p«ple). On the 
other hand the French beaute, which 
was called {hemwiee, heoiee) is here 
written beivte, which was clearly 
(bEutf^). Similarly German words, as 
knew, cannot have been anything but 
(kneo, kuEu). Similarly (uEue) new. 
The French diphthong oi as in vois 

Khbhan that ^prille bhith His shuures soot 
The drukht of martsh HOth pErsed too the 

^nd brtthyd Evri veeii in sbhitsh likuur 
Of khbhitsh vertyy- KndzhEndred is the 

fiuur, 4 

Khbhon Scfirys ee'k bhith His sbheete breeth 
Enspiiryd Hath in evi-I hoU and neeth 
The tEndre kroppes, and the joqge sonne 
Hath in the Ram His Halfe kurs ironne, 8 
-4nd smale fuules maken melodiie 
That sl^^pen a\ the nikht bhith oopen iie, 
Soo priketh heoi natyyr- in HEr koradzhes, 
Than loqgen folk too goon on pilgrimadzhes, 
.4nd palmers for too seeken strAAndzhe 

strondes 13 

Too fErne nalbhes, kuuth- in sondri londes, 
And spesialli from Evri shiires Ende 
Of Eqglond too Kontyrbyri thee bhEnde 16 
The Hooli blissfyl martir for too seeke 
That HEm nath Holpen khbhan that thee 

bheer seeke. 
BifEll that in that sesuun on a dEE 
In Suuth-bhErk at the tabbard as ii Iee, 20 
Beedi too bhEnden on mii pilgrimadzhe 
Too Kaentyrb-ri bhith fyl devuut koradzhe, 
At nikht bhas kom intoo that hostelriie 
BhEl niin and tbhsnti in a kompaniie 24 
Of sondri folk bii aventyyr- ifalle 
In fElaship, and pilgrims bheer bhi alle 
T^at tobhard Kantyrbyri bholden riide. 
The tshambers and the stables bheeren 

bhiide. 28 

voice, was taken over unaltered, and 
also replaces romanic ui, which was 
too far removed from English feelings ; 
we have seen fruit pass into (fryyt, 
fruut) ; ennuyer becomes (anoi) and 
destruire is written dentruie, deatrie, 
but had the same sound (destrai). 

As regards the so-called mute «, it 
was undeniably historical in Chaucer 
and represented old inflections, yet it 
was, with equal certainty, in many 
cases merely mechanically imitated 
from the French. But we cannot scan 
Chaucer in the French fashion, with- 
out omitting or inserting the mute e at 
our pleasure, and in a critical edition 
of the poet, the spoken e only ought to 
be written. "What was its sound when 
spoken ? Certainly not (a) as in 
French, but a pure (e) with some in- 
clination to (i) . This is shewn by the 
rh}Tne (soothe, too ihee) already cited, 
and many others, as clerkes, clerk is\ 
(dr<?cd is, deedes) etc. At present 
Englishmen pronounce this final e in 
the same way as i, and in general e,i 
present as natural a euphonicum as the 
French (a). 

The following are the opening lines 
of the Canterbury Tales reduced to a 
strict metre. 

[Some misprints seem to occur in 
the original, but I have left them un- 

^nd bhEl bhe bheeren eesyd atte bEste, 
AnA shortli khbhan the sonne bhas too reste 
Soo Had ii spoken bhith hehi Evritsh-oon 
That ii bhas of HEr fElaship anoon 32 

^nd mAAde forbhard Erli too ariise 
Too tak- uur bhEE thEr as ii juu debhiise, 
Byt nAAthelESS, khbhiils ii nabh tiim and 

Or that ii fErther in this tale pase 36 

Me thiqketh it akordant too resuun 
Too telle juu all the kondisiuun 
And. khbhitsh thee bheeren and of khbAat 

Of eetsh of HEm, soo as it seemed mee 40 
And. eek in khbhat crrEE that thee bhcer- 

^nd ot a knikht than bhol ii first beginne. 
A knikht thEr bhas and that a bhorthi 

That from the tiime that ne first bigan 44 
Too riiden uut He loved tshivalriie 
Truuth and Honuur, freedoora and kyi-tesiie. 
Fyl bhorthi bhas ne in His lordes bhErre 
^nd thErtoo nadd He riden nooman fErre 48 
As bhEl in kristendoom as neethenEsse 
^nd Ever Honuurd for His bhorthinEsse. 
Ai Alisandr- He bhas khbhan it bhas bhonne, 
Fyl ofte tiim He nadd the bord bigonne 52 
^boven alle nasiuuns in Pryse, 
In Lettoou nadde rEEsed and in Ryse 
Noo kristen man soo oft of nis degree, 
In GErnad- alte siidzhe nadd He bee, 56 


At mortal batEEls nadd ne been fiifteene 61 Bhith lokkes kryll- as thee bhEr lEEd in 
^nd fukhten for uur fEEth at Tramasseene, prEsse, 

In listes thriies and ee slEEn His too. Oi tbhEnti jeer He bhas of adzh- ii gesse, 

This ilke bhortbi knikht nadd been alsoo 64 Oi His statyyr- ne bhas of Even lEqthe 83 

Somtiime bhitA the lord of Palatiie ^nd bhondyrli delivr- and greet of strEqthe, 

jigEEn another neethen in Tyrkiie, -4nd He h«dd been somtiim in tshivatshiie 

^nd Evermoor ne Hadd a sovrEEn priis. In Flandres, in Artois and Pikardiie, 

^nd thukh that He bhas bhorthi ne bhas And born nira bhEl, as in soo litel spase 

bhiis, 68 In nop too stonden in his ladi grase. 

^nd of His port as miik as is a meed. Embruudid bhas ne as it bheer a raEde 88 

He nEver jit a vilonii ne sEEd A\ fyl of frEshe fluures, khbhiit- and reede. 

In al His liif, yntoo noo maner bhikht. Siqgiqg ne bhas or fluutiqg al the (/ee, 

He bhas a vErrEE pErfikht dzhEntil knikht. He bhas as fresh as is the moonth of mEE, 92 

Byt for too tElle juu of His arrEE, 73 Shor< bhas His guun bhith sleeves loqg and 

His Hors bhas good, byt He ne bhas nukht bhiide, 

gEE, BhEl kuud He sitt- on Hors and fEEre riide, 

Of fystian He bhsred a dzhepuun He kuud soqges bhEl make and endiite, 

A\ bismoteryd bhith His naberdzhuun, 76 Dyhystn- and eek dAAns- ond bhEl pyrtrsE 
For He bhrts lat komen from His viadzhe and bhriite. 96 

J.nd bhsnte for too doon his pilgrimadzhe. Soo Hoot ne lovde, that bii nikhter-tale 

Bhith Him thsr bhas his son, a Joqg He sleep nomoor than dooth a nikhtiqgale. 

skbhieer, KyrtEES He bhas, lukhli [or loouli) and 
A lovjer and a lysti batshelcer 80 SErvisable 

^nd karf beforn His fadyr at the table. 100 

If in the above we read (ee, e) and (oo, o) for {ee, e) and (oo, o), 
and (e) for (e) which is a slight difference, and also («V, i) for (ii, i), 
and do not insist on (a) for (a), and also read (w, wh) for the un- 
English (bh, khbh), the differences between this transcript and 
my own, reduce to 1) the treatment of final e, which Eapp had not 
sufficiently studied ; 2) the merging of all short u into (y), certainly 
erroneous ; 3) the indistinct separation of the two values of ou into 
(uu, oou), and 4) the conception of (ee), an un-English sound, as 
the proper pronunciation of ey, ay as distinct from long e. It is 
remarkable that so much similarity should have been attained by 
such a distinctly different course of investigation. 

Instructions for Reading the Phonetic Transcript of the Prologue. 

The application of the results of Chapter IV. to the exhibition 
of the pronunciation of the prologue, has been a work of great 
difficulty, and numerous cases of hesitation occurred, where analogy 
alone could decide. The passages have been studied carefully, and 
in order to judge of the effect, I have endeavoured to familiarise 
myself with the conception of the pronunciation by continually 
reading aloud. The examination of older pronunciation in Chap. 
v., has on the whole confirmed the view taken, and I feel con- 
siderable confidence in recommending Early English scholars to 
endeavour to read some passages for themselves, and not to pre- 
judge the effect, as many from old habits may feel inclined. As 
some difficulty may be felt in acquiring the facility of utterance 
necessary for judging of the effect of this system of pronunciation, it 
may not be out of place to give a few hints for practice in reading, 
shewing how those who find a difficulty in reproducing the precise 
sounds which are indicated, may approximate to them sufficiently 
for this purpose. These instructions correspond to those which I 
have given in the introduction to the second edition of Mr. R. 
Morris's Chaucer. 

The roman vowels (a, e, o, u) must be pronounced as in Italian, 


with the broad or open e, o, not the narrow or close sounds. They 
are practically the same as the short vowels in German, or the 
French short a, e, o, ou. The (a) is never our common English a in 
fat, that is (a?), but is much broader, as in the provinces, though 
Londoners will probably say («). For (o) few will perhaps use 
any sound but the familiar (o). The (u) also may be pronounced 
as (w), that is, u in hull or oo in foot. The long vowels are 
(aa, ee, oo, uu) and represent the same sounds prolonged, but if 
any English reader finds a difficulty in pronouncing the broad and 
long (ee, oo) as in Italian, Spanish, Welsh, and before r in the 
modern English mare, more, he may take the easier close sounds 
{ee, oo) as in male, mole. The short {i) is the English short i in 
pit, and will occasion no difficulty. But the long (n) being un- 
usual, if it cannot be appreciated by help of the directions on p. 
106, may be pronounced as (ii), that is as ee in feet. The vowel 
(yy), which only occurs long, is the long French u, or long German 
il. The final (-e) should be pronounced shortly and indistinctly, 
like the German final -e, or our final a in China, idea, (supra p. 119, 
note, col. 2), and inflectional final -en should sound as we now pro- 
nounce -en in science, patient. It would probably have been more 
correct to write ('b) in these places, but there is no authority for 
any other but an (e) sound, see p. 318. 

For the diphthongs, (ai) represents the German ai, French, al 
Italian ahi, Welsh ai, the usual sound of English aye,'^ when it is 
distinguished from eye, but readers may confound it with that 
sound without inconvenience. The diphthong (au) represents the 
German au, and bears the same relation to the English ow in now, 
as the German ai to English eye, but readers may vrithout incon- 
venience use the sound of English ow in now. Many English 
speakers habitually say (ai, au) for (oi, au) in eye, now. The diph- 
thong (ui) is the Italian ui in lui, the French out nearly, or more 
exactly the French oui taking care to accent the first element, and 
not to confound the sound with the English we. 

The aspirate is always represented by (H h), never by (h), which 
is only used to modify preceding letters. 

(J j) must be pronounced as German j in ja, or English y in yea, 
yawn, and not as Englishy in just. 

The letters (b d fgk Imnprstvwz) have their 
ordinary English meanings, but it should be remembered that (g) 
is always as in gay, go, get, never as in gem ; that (r) is always 
trilled with the tip of the tongue as in ray, roe, and never pro- 
nounced as in air, ear, oar ; and also that (s) is always the hiss in 
his5 and never like a (z) as in hi^, or like (sh). The letter (q) has 
altogether a new meaning, that of ng in sing, singer, but ng in 
finger is (qg). 

^ This word is variously pronounced, text is generally used in the South of 

and some persons rhyme it with nay. England, but this pronunciation is per- 

In taking votes at a puhlic meeting the haps unknown in Scotland, 
sound intended to be conveyed in the 


(Th, dh) represent the sounds in thin, then, the modem Greek B- 

(Sh, zh) are the sounds in me^A meaifure, or pisA, vision, the 
Fr. ch,j\ 

(Kh, gh) are the usual German eh in ach and ff in Ta^e. But 
careful speakers will observe that the Germans have three sounds 
of ch as in ich, ach, Siuch, and these are distinguished as {Jch, kh, 
'kwh) ; and the similar varieties (^h, gh, gwh) are sometimes found. 
The reader who feels it difficult to distinguish these three sounds, 
may content himself with saying (kh, gh) or even (h'). The (k.wh) 
when initial is the Scotch quh, Welsh chw, and may be called 
(khw-) without inconvenience. Pinal {gwh) differs little from 
(wh) as truly pronounced in when, whsit, which should, if possible, 
be carefully distinguished from (w). As however (wh) is almost 
unknown to speakers in the south of England, they may approxi- 
mate to it, when initial, by saying (h'u), and, when final, by 
saying (uh'). 

The italic (w) is also used in the combination (kw) which has 
precisely the sound of qu in queen, and in {vw) which may be pro- 
nounced as (rw), without inconvenience. 

(Tsh, dzh) are the consonantal diphthongs in chest jest, or such 

The hyphen (-) indicates that the words or letters between which 
it is placed, are only separated for the convenience of the reader, 
but are really run on to each other in speech. Hence it frequently 
stands for an omitted letter (p. 10), and is frequently used for an 
omitted initial (h), in those positions where the constant elision of 
a preceding final -e shews that it could not have been pronounced 
(p. 314). 

These are all the signs which occur in the prologue, e